The in-depth guide FORENSIC SCIENCE SEMINAR for forensic professionals Forensic sci. sem. The best professional researches from global experts. Articles Reviews News 1 Scene 1 Investigation 1 Conferences 2 Laboratory 2 Identification 2 Cases 3 Simulation 3 Interdisciplinary 3 Trainings

Z o l C a t A c a d e m i c FSSISSN 2157-118X H o u s e Editor-in-chief: Tilla A. Theresia Issue 14 Vol.8 No. 1 w w w . z o l c a t . c o m C a l l f o r April 2018 http://fss.xxyy.info P a p e r s

FORENSIC SCIENCE SEMINAR PUBLISHED BY ZOLCAT ACADEMIC HOUSE

NEW YORK ISSN 2157-118X 3 April, 2018, Volume 8, Number 1 Forensic Science Seminar

Contents

Review Articles

1 Application Status and Intelligent Trend of Digital Forensic Pathology in Identification Practice Qi GUO, Haibin XUE

5 A Generic Process Model for Botnet Forensic Analysis

Meenakshi Thapliyal, Anchit Bijalwan, Neha Garg, Emmanuel Shubhakar Pilli

10 Forensic Accounting on Corporate Governance Maturity mediated by Internal Audit: A Conceptual Overview Ali Rehman, Fathyah Hashim

Original Research Papers 19 Two different approaches to handle landmark location uncertainty in skull-face overlay:coevolution vs fuzzy landmarks Óscar Ibáñez, Óscar Cordón, Sergio Damas Single Issue Price $ 000 USD (Free / Non-commercial) Peer Reviewed | Limited Open Access 27 Deceleration Impact Experiments and Deceleration Injury Mechanism Analysis on the Thoracic and Abdominal Organs Editor-in-Chief: Tilla A. Theresia, Ph.D. - Quality Hongfa Chang, Daiqin Tao, Shengxiong Liu, Jingru Xie, Guangyu and Technology Management on Forensic Science Yang, Zhiyong Yin, Zhengguo Wang Laboratory (USA). E-mail: [email protected] Editorial Board 32 Effects of Heat Exposure Duration on Saliva Traces on Tarri Querina Borbón Zóe de Llaneadora, Ph.D. - Cigarette butts as Forensic Identification Tools Criminal Scene (Spain); Prof. Eric YIN, Ph.D. - Biomechanical Forensics (China); Prof. Donald A. Mely Purnadianti, Andika Aliviameita, Diah Ayu Nur Rochmawati, Redelmeier , - Traffic Medicine (Canada); Assoc. Dian Amanovitasari Prof. Eric ZENG, Ph.D. - Structure Safety (China); Quiéta de Thérésa, Ph.D. - Forensic Chemistry (France); Kyu Hana Daun, Ph.D.- Optics and Corrigendum Software Engineering ( Japan); Prof. Mark Benecke, M.D. - Forensic Biology (Germany); Галина C3 Errata: Corrigendum and Apology

Сергеевна Мысливцев, Ph.D. - Criminal Editorial Psychology (Russia)

Forensic Science Seminar ISSN 2157-118X Editorial Department E-mail: [email protected] Official Website http://fss.xxyy.info/ Publisher ZolCat Academic House, Brooklyn, NY, 11220. Indexed by Google Scholar, OCLC (659514459), Ulrich. Impact Factor = 0.179 (2017). Sponsored by the Library of Congress. FORENSIC SCI SEM Printed in the United States of America. All Rights Reserved. © 2011-2018.

C2 GUO Qi, XUE Haibin. Application status and intelligent trend of digital forensic pathology in identification practice. Forensic Sci Sem, 2018, 8(1): 1-4.

Peer Reviewed Forensic Science Seminar Received 1 February 2018 ISSN 2157-118X Received in revised form 6 March 2018 Volume 8 Number 1 Accepted 7 April 2018 3 April 2018 Available online 7 May 2018

pplication Status and Intelligent Trend of Digital AForensic Pathology in Identification Practice GUO Qi a, XUE Haibin b, *

a. Public Security Bureau of Bishan District in Chongqing, Chongqing, 400000, China. b. Liangjiang New Area Branch of Chongqing Public Security Bureau, Chongqing, 400000, China. *CORRESPONDING AUTHOR. E-mail: [email protected]; Tel: +86-13983045601.

ABSTRACT The traditional forensic identification methods rely mainly on the naked eye observation and empirical judgment of the appraisers in the process of autopsy and histopathological examination. The impartiality and scientific nature of the identification opinions thus formed are often questioned, often causing the parties to suffer due to evidence. Or conflict with the case handling party. Forensic pathology and medical imaging, computer simulation and other multidisciplinary integration formed a series of new digital technologies, which are revolutionizing forensic identification technology. KEY WORDS Forensic pathology, Virtual anatomy, Multi-rigid body simulation, Finite element analysis, FEA, Forensic science

the body image data, is conducive to late re-examination and 1. INTRODUCTION consultation, the detection of forensic injury and the cause of death Forensic autopsy is a fundamental method and gold standard for has a great value. [2] recognizing clear causes of death. The traditional forensic In the 1990s, Dr. Thali et al. of the Institute of Forensic identification methods rely mainly on the naked eye observation Medicine at the University of Bern, Switzerland, first proposed the and empirical judgment of the appraisers in the process of autopsy concept of virtual anatomy and formed a theoretical system. and histopathological examination. The impartiality and scientific Current virtual anatomy techniques include optical nature of the identification opinions thus formed are often three-dimensional surface scanning, 3D photography, CT scans, questioned, often causing the parties to suffer due to evidence. Or and MRI scans. They make full use of the advantages of new conflict with the case handling party. Some of the lesions may not technologies and multi-technology integration to address causes of be able to obtain enough powerful information through autopsy, death, forensic pathology, life response, injury reconstruction, such and the traditional autopsy is a destructive examination. For as forensic focus. Many countries in Switzerland, the United States, centuries, it has been resisted and opposed by the families of the Japan, the United Kingdom, and Australia [2-5] have achieved deceased, especially by some religious cultural traditions [1]. certain research results in terms of damage detection, injury model Forensic pathology and medical imaging, computer simulation estimation, and cause of death identification. and other multidisciplinary integration formed a series of new 3. MULTI-RIGID BODY SIMULATION digital technologies, which are revolutionizing forensic identification technology. With the development of economy, the increase in the types and At present, virtual anatomy, multi rigid body simulation and number of vehicles, the incidence of traffic accidents and the finite element analysis are the most mature technologies. number of dead and injured have also shown an upward trend, which has become an important factor affecting social development 2. VIRTUAL ANATOMY and people’s lives. The identification of traffic accidents has also With the development of medical imaging techniques such as CT gradually risen to a forensic pathology, the focus identification and MRI, it provides a non-invasive/minimal invasive cadaveric project. In recent years, with the development of computer examination method for forensic medical examination—the hardware and numerical simulation technology, it has become forensic virtual anatomy technique that can clearly observe the technically possible to reconstruct traffic accidents through digital cadaver organs, bone lesions and damage, and can save the parts of reconstruction methods. The digital photogrammetry technology

1 GUO Qi, XUE Haibin. Application status and intelligent trend of digital forensic pathology in identification practice. Forensic Sci Sem, 2018, 8(1): 1-4. and multi-rigid body dynamics simulation technology can identification of injury. reconstruct the accident scene and digitally simulate, simulate and Since the 1970s, the finite element method has been applied to visualize the collision and injury process in the accident, thus the modeling and damage biomechanics of the human head, spine, assisting in the development of the forensic identification of traffic etc. The research scope has gradually expanded to the pelvis, injuries, and the accuracy, objectivity of accident identification, and thorax, and joints of the limbs and other structures. In addition to impartiality and scientific treatment of accident liability. the skeletal system, finite element models have been constructed Since the 1970s, research institutes in the United States, Europe, for cardiovascular, ocular, gastrointestinal, pulmonary, hepatic, Japan, and Australia have used multi-rigid body dynamics methods renal, and even placental sites and lesion analyses have been to perform vehicle crash tests and impact tests of occupant performed. At present, finite element modeling and analysis protection devices, which have caused structural deformation and methods have achieved certain results in the following crashworthiness of vehicles in collision accidents. As well as issues biomechanical studies of common injuries, including accelerations such as safety, a simulation analysis was carried out to reveal the caused by acceleration/deceleration of the head and injuries caused injury mechanism of the vehicle's personnel, formulate damage by rotational motion, and internal organs caused by the external assessment standards, and propose improvements to the vehicle force of the chest and abdomen, injury to blood vessels and damage design. Since the beginning of the 21st century, European and to the thorax, spine, pelvis, and joints of the limbs and attachments American research institutions have begun to use multi-rigid caused by external violence. [14-18] pedestrians and vehicle models to reconstruct real road traffic The study on the relationship between finite element modeling accidents, simulate car-pedestrian collisions, reproduce the process and human injury started relatively late in China. In recent years, of injury caused by traffic accidents, and determine the posture of there have been reports of research results in succession, mainly pedestrians, the collision speed and the braking state of vehicles, focusing on the fields of engineering and material mechanics, etc. And analyze the impact of factors such as vehicle type, body traffic injury prevention and clinical trauma research. The Southern shape, initial posture, and vehicle braking status on pedestrian Digital Medical University, the Third Military Medical University, distance, location and human injury after collision. [6-9] and Shanghai Jiao Tong University have developed digital avatar The research on simulation and simulation of multi-rigid traffic programs for medical research, clinical treatment and teaching, and accidents in China started relatively late. The digital reconstruction have achieved corresponding research results. The human of car-pedestrian crashes was mainly carried out after 2002. biomechanical model research plan is used to simulate the collision Tsinghua University, Jilin University, Hunan University, Third between different parts of the vehicle and the human body in a Military Medical University and other research institutes have traffic accident, to perform numerical response analysis and to independently constructed or apply commercial models to study the injury mechanism of human injury. [19-20] In addition, reproduce the real-world case of car-pedestrian collision accidents. based on the research needs of weapon damage protection; relevant The research content includes the relationship between vehicle agencies in China have also conducted relevant research on chest speed and body throw distance. The characteristics of different finite element modeling and damage analysis. [21-22] vehicle speeds and different vehicle models causing damage to the However, there is a certain gap between the focus of the above outside of the car are discussed, including the possibility of research work and the damage morphology, damage mechanism, protection from outside the vehicle. [10-13] and biomechanical analysis of damage in forensic science research.

4. FINITE ELEMENT ANALYSIS (FEA) 5. DISCUSSION

The analysis of human forensic pathology by traditional forensic Through virtual anatomy technology, human injury can be clearly pathology mainly depends on naked eye observation, autopsy, light detected and the damage situation can be effectively evaluated. microscope observation and empirical judgment. The subjective Through multi-body dynamics simulation technology, road traffic arbitrariness is strong, and the understanding of damage is mostly accident reconstruction can be effectively carried out, and the limited to damage morphology. [14] occurrence of real accidents and the damage of human injuries can The finite element model of the human body is composed of the be restored. Through finite element modeling and analysis decomposition and reorganization of human tissues in unit discrete techniques, the damage can be dynamically and quantitatively and three-dimensional modeling. It can perform load simulation on analyzed from the perspective of biomechanics to predict the site of specific boundary conditions, calculate the effect of external forces injury and observe the occurrence and development of the injury. on human tissue, and finally obtain various Human biomechanical The formed electronic evidence can be displayed in the form of results. The finite element method is of great help in reconstructing images and dynamic effects, clear, easy to understand, easy to and predicting the locations, forms, and degrees that may occur in understand the illegal medical personnel involved in the litigation human injury, improving the shortcomings of traditional forensic process, and greatly enhance the quality of the identification of the pathology, and providing a favorable reference for forensic case. In view of the differences in the principles, equipment, and

2 GUO Qi, XUE Haibin. Application status and intelligent trend of digital forensic pathology in identification practice. Forensic Sci Sem, 2018, 8(1): 1-4. concerns of the various digital technologies mentioned above, the more than 50 years old for the oldest technologies. On the other joint application of multiple digital technologies to complement hand, several issues that are closely related to the nature of big data each other has become a research hotspot and development trend in and inherited from other scientific fields (such as artificial the international forensic science field. [23] intelligence) are often underestimated if they are not ignored. In According to relevant reports [24], the development of various addition, some papers have adjusted almost identical enthusiasm new technologies in forensic pathology digitalization still has for big data, and because they describe the risks, they deserve deficiencies, and each has its own advantages over traditional attention. [30] In this context, forensic science is still waiting for its forensic identification. The combination of traditional forensic position paper and a comprehensive overview of what big data can identification and digital technology can provide a more scientific bring in this area. The current situation requires that definitions and and effective theoretical basis for forensic identification. actions reasonably guide the research and practice of big data. This The preliminary results based on the concept of "virtual reality", is an opportunity for true interdisciplinary cooperation. such as autopsy multi-slice spiral computed tomography (MSCT) REFERENCE and magnetic resonance imaging (MRI), have been introduced and evaluated in forensic medicine [25]. Forensic pathology new 1. Lundberg GD. Low-tech autopsies in the era of high-tech medicine: continued value for quality assurance and patient safety [J]. JAMA, molecules and innovative technologies are also in full swing [26]. 1998, 280: 1273-1274. While some research teams have recently begun to use artificially 2. Thali MJ, Dirnhofer R, Vock P. The virtopsy approach[M]. CRC related technologies to conduct new research on forensic pathology, Press，2009. 3. Buck U, Christe A, Naether S, et al. Virtopsy -- Noninvasive for example, to improve the quality of autopsy reports through Detection of occult bone lesions in postmortem MRI: Additional computer proofreading procedures to identify errors [27]; to vector information for traffic accident reconstruction [J]. Int J Legal Med, [28] 2009, 123(3): 221-226. quantifiers to identify endometrial malignancies ; to establish a 4. Buck U, Naether S, Braun M, et al. Application of 3D documentation diagnostic decision support system through logical programming, and geometric reconstruction methods in traffic accident analysis: with high resolution surface scanning, radiological msct/mri scanning supplemented by a computational framework based on an artificial and real data based animation [J]. Forensic Sci Int, 2007, 170(1): neural network, establishing a model for assessing thrombophilic 20-28. propensityshment of a diagnostic decision support system 5. Dirnhofer R, Jackowski C, Vock P, et al. VIRTOPSY: Minimally invasive, imaging -guided virtual autopsy [J]. Radiographics, 2006, supplemented by formal agendas based on logical programming 26(5): 1305-1333. methods for knowledge representation and reasoning, 6. XU Jiong. Research and application of motor-bus collision accident reconstruction oriented to occupant injury [D].Shanghai: Shanghai supplemented by a computational framework based on artificial Jiao Tong University, 2010. (In Chinese) neural networks, and its model is quite accurate in assessing 7. SHEN Jie. Research and application of vehicle-pedestrian accident reconstruction [D]. Shanghai: Shanghai Jiao Tong University, 2007. propensity to thrombogenicity. This model can properly classify (In Chinese) patients who do present with pathology and can classify disease 8. Untaroiu C D, Crandall J R, Takahashi Y, et al. Analysis of running child pedestrians impacted by a vehicle using rigid-body models and deletions [29]; etc. optimization techniques [J]. Safety Science, 2010, 48(2): 259-267. 9. Untaroiu C D, Meissner M U, Crandall J R, et al. Crash 6. CONCLUSION reconstruction of pedestrian accidents using optimization techniques[J]. International Journal of Impact Engineering, 2009, At present, forensic pathology digitization technology is still in the 36(2): 210-219. development stage. Virtual anatomy, multi-rigid body simulation, 10. BAI Zhonghao, CAO Libo, YU Zhigang. A research on the difference of frontal impact response between 50th percentile finite element analysis and other technologies all have certain flaws Chinese male and hybrid III 50th percentile male [J]. Automotive and technical blind spots. Digital technology cannot completely Engineering, 2008, (11): 993-997, 1005. (In Chinese) 11. CHEN Yong. Research on protection of child-pedestrian head injury replace traditional forensic identification and research methods. in vehicle-pedestrian collisions [D]. Hunan: Hunan University, 2008. Combining digital technology with traditional identification (In Chinese) 12. FENG Cheng-jian, WANG Fu-ping, XU Chen, et al. Head dynamic methods, a new injury analysis model including expert system, data response based on reconstruction of vehicle-pedestrian accidents with information system, and computer system was formed to ensure the the video [J]. Journal of Medical Biomechanics, 2013, 28(2): forensic medicine and the rigor, objectivity, and accuracy of 164-170. (In Chinese) 13. GU Jie, XIA Yong, ZHOU Qing. Kinematics of the head in scientific evidence formed. cyclist-vehicle and pedestrian-vehicle collisions [J]. J Automotive In less than a decade, big data in medicine has become a Safety and Energy, 2014, 5(2): 159-165. (In Chinese) 14. LIU Ningguo, CHEN Yijiu. The theory and practice of new digital phenomenon, and many biomedical disciplines have their own techniques of forensic pathology [M]. Shanghai SciTechEducation papers on this topic. Outlook and debate are booming. However, Press, 2015. (In Chinese) 15. Jean R, Caroline D, Rémy W, et al. Finite-element models of the real big data research is still very rare in medicine and does not human head and their applications in forensic practice [J]. Int J Legal meet all expectations. On the one hand, technologies that are Med, 2008, (122): 359-366. 16. Raul J S, Baumgartner D, Willinger R, et al. Finite element usually proposed for big data (such as machine learning techniques) modelling of human head injuries caused by a fall [J]. Int J Legal should support ambition for personalized, predictive and preventive Med, 2006, 120(4): 212-218. drugs. Most of these technologies are not new, and they have been 17. El-Rich M, Arnoux P J, Wagnac E, et al. Finite element investigation of the loading rate effect on the spinal load-sharing changes under

3 GUO Qi, XUE Haibin. Application status and intelligent trend of digital forensic pathology in identification practice. Forensic Sci Sem, 2018, 8(1): 1-4.

impact conditions [J]. Journal of Biomechanics, 2009, 42(9): 1252. perspective for forensic science [J]. The American Journal of 18. Li Z, Kindig M W, Kerrigan J R, et al. Rib fractures under anterior Forensic Medicine and Pathology: official publication of the National -posterior dynamic loads: experimental and finite-element study [J]. Association of Medical Examiners, 2011, 32(3): 280-286. Journal of Biomechanics, 2009. 25. Thali M J, Yen K, Schweitzer W, et al. Virtopsy, a new imaging 19. ZHOU Ze-min, FANG Chi-hua, HUANG Li-wei, et al. The study of horizon in forensic pathology: virtual autopsy by postmortem pancreas three-dimensional reconstruction based on the Virtual multislice computed tomography (MSCT) and magnetic resonance Chinese Human-Female No 1 [J]. Chin J Surg, 2005, 43(21): imaging (MRI)-a feasibility study [J]. Journal of forensic sciences, 1401-1404. (In Chinese) 2003, 48(2): 386-403. 20. HE Pei. The development of a three-dimensional finite element 26. Kondo T, Ishida Y, Kimura A, et al. New Molecular and Innovations model of the human brain and frontal impact simulation and analysis in Forensic Pathology[M]//P5 Medicine and Justice. Springer, Cham, [D]. Tianjin: Tianjin University of Science Technology, 2006. (In 2017: 126-145. Chinese) 27. Cain M D, Brazelton J, Dye D W. Identifying errors in forensic 21. LI Hongbo. Finite element modeling and simulation of human thorax autopsy reports using a novel web-based program [J]. Academic protection based on behind armor blunt trauma [D]. Luoyang: Henan Forensic Pathology, 2016, 6(1): 103-108. University of Science and Technology, 2012. (In Chinese) 28. Pouliakis A, Margari N, Karakitsou E, et al. Artificial intelligence for 22. ZHANG Zhigang. Research of human thorax FEM for analysis of the identification of endometrial malignancies: Application of the explosion protection [D]. Chongqing: Third Military Medical learning vector quantizer [J]. International Journal of Reliable and University, 2008. (In Chinese) Quality E-Healthcare (IJRQEH), 2018, 7(2): 37-50. 23. CHEN Yi-jiu. The development and application of new digital 29. Vilhena J, Vicente H, Martins M R, et al. An artificial intelligence techniques in forensic pathology [J]. Chinese Journal of Forensic approach to thrombophilia risk [J]. International Journal of Reliable Sciences, 2016, (3): 64-71. (In Chinese) and Quality E-Healthcare (IJRQEH), 2017, 6(2): 49-69. 24. Giovanna Sansoni, Cristina Cattaneo, Marco Trebeschi et al. Scene 30. Lefèvre T. Big data in forensic science and medicine [J].Journal of of crime analysis by a 3-dimensional optical digitizer a useful Forensic and Legal Medicine, 2017, In Press (Corrected Proof).

POSTSCRIPT N/A

4 Meenakshi Thapliyal, Anchit Bijalwan, Neha Garg, et al. A generic process model for botnet forensic analysis. Forensic Sci Sem, 2018, 8(1): 5-9.

Peer Reviewed Forensic Science Seminar Received 3 November 2017 ISSN 2157-118X Received in revised form 26 December 2017 Volume 8 Number 1 Accepted 4 January 2018 3 April 2018 Available online 2 April 2018

Generic Process Model for Botnet Forensic A Analysis Meenakshi Thapliyal a, *, Anchit Bijalwan b, Neha Garg a, Emmanuel Shubhakar Pilli a

a. Department of Computer Science and Engineering, Graphic Era University, Dehradun, Uttarakhand, India. b. Department of Computer Science and Engineering, Uttarakhand Technical University, Dehradun, Uttarakhand, India. *CORRESPONDING AUTHOR. E-mail: [email protected]

ABSTRACT Botnets are becoming more hazardous in cyber crime when compared to other malicious activities. Security against botnets is a major concern. Botnet forensics is young science which can answer questions about how, what and where of damage done by bots. The forensic system deals with capturing, recording, and analysis of botnet traffic. This paper outlines the process of Botnet forensic analysis and its implementation. A generic process for botnet forensics is proposed based on previous digital forensics models. The specific research gaps existing in implementation are identified and presented as challenges. The contribution of this work is that it presents an overview on botnet forensics analysis and implementation which will be more valuable for security. KEY WORDS Bot, Botnet, Analysis, Forensic science

forensics can be used to evaluate how and where the attack 1. INTRODUCTION occurred, who was culprit, duration of the exploit, and the line of Botnet is an army of infected computers that take instructions from attack. Botnet forensics can be used as a device for monitoring the a botherder. A botherder is corrupt hacker who uses the botnet for activity of botmaster and hacker, business transaction analysis and financial gain or as a destructive behavior of civilization and the investigation of irregular performance issues sources. Botnet Internet community without ethics. The bad guys have been using forensics involves postmortem research of the attack means notitia the latest killer web application, is the advanced security Web criminis (after crime announcement) which takes particular technology. Many of the security professionals who pioneered the duration of action and way of tackle for particular cases. Botnet fight against botnets are demoralized by the realization that taking forensics and the attacker both are at the identical proficiency out the C&C [1] does not helps. Botnets a call to Action (Command altitude. The hacker uses a set of tools to launch the attack and the and Control) server is less effective now days. Botnet infection is botnet forensic authority applies similar tools to explore the attack. adaptive means for chain to chain system. Single virus/worm The hacker has all the time at his disposal and will regularly get spread the infection another module through malicious code that better his skills, provoked by the millions of dollars in resolving prevents previous antivirus action [2]. risk. The paper is structured as follows: in Section 2 introduces the Social networking sites like orkut, facebook, skype and Google botnet forensic methodology and motivation, Section 3 brief blogger were infected by distributed denial of service attacks, spam summary of previous botnet forensic analysis. We propose a etc. Computer forensics was introduced by law enforcement with generic process model for botnet forensics analysis. Key proper guidelines of judicial system [3]. Computer forensics Challenges are presented in Section 4, conclusion is given in involves protection, detection, mining, records, and analysis of Section 5. computer data. Botnet forensics analysis is a natural extension of 2. BACKGROUND computer forensics. Botnet forensics analysis is a permanent monitoring process that deals with capturing, recording, and Internet Relay Chat (IRC) is a text-based chatsystem that analysis of botnet traffic in packets format. It can also indicate communicates bot in channels. They were capable to interpret alerts when thresholds (maximum limit) are exceeded. If the simple commands, provide administration support, suggest simple present attack could not be prevented, the fundamental information games and regain information about operating systems, logins, is used to defend against similar attack in future event. Botnet email addresses, aliases, etc. [4]. Denial of service, then distributed

5 Meenakshi Thapliyal, Anchit Bijalwan, Neha Garg, et al. A Genericgeneric process model for botnet forensic analysis. Forensic Sci Sem, 2018, 8(1): 5-9. denial of service attacks were implemented in these IRC bots. They architecture, and describing a behavioral model, for a specific class could proliferate like worms, remain hidden as viruses and could of network users, able to identify botnet-related activities. Time initiate huge, corresponding attacks like AgoBot and SDBot. The analysis patterns of botnet activity and taxonomy both phenomenon recent generation of bots can multiply through filesharing networks, are established through Botnet traffic analysis. This IRC traffic peer-to-peer networks, email attachments and infected websites. analysis is a versatile approach which based on systematic Bots communication can be accomplished by several protocols, monitoring, for botnets detection which based on the network such as IRC, HTTP and P2P. Latest sophisticated botnet is CoolBot, behavior. Initially analysis has been supervised then next testing of which could explain the problems that how to recover shut down pure invalid anomaly was detected. Kugisaki et al. [8] studied on and accept delay problem of C&C and contradict against insecurity IRC behavior which directly related to an IRC server. This novel (routing table poisoning). CoolBot could control C&C model approach can be used to make judgments between of present bots automatically and repair the broken C&C, which means C&C more than existing viruses by computerization. More objects [5] reconstruction . (problem of verification and generality of objecs), duration of 2.1. Botnet forensic methodology detection by real visible and Measures against bots that does not Botnet forensic methodology consists of three steps: use IRC. Malware collection 3.2. Asprox botnet analysis A ‘Catch-it-as-you-can’ system follows where all bots packets Borgaonkar et al. [9] studied the design and structure of the passing through an exacting traffic point are captured. Analysis is Asprox botnet, the communication protocols which drive-by consequently done which require huge storage. A piece of bots downloading of spreading malicious substance and the advanced packet is individually analyzed and stored in definite memory for fast-flux service network. The main features of the Asprox botnet future analysis [6]. are the use of centralized command and control structure (C & C Malware analysis structure), HTTP based communication, advanced double fast-flux The botnet forensic system is a security device with hardware service networks, SQL injection attacks for recruiting new bots and and pre-installed software. It is defined behavior-based analysis. social engineering tricks to spread malware binaries. Hydra Botnet tracking fast-flux network, SQL injection attack tool is advanced features of The extracted information from malware collection interferes Asprox. Asprox botnet does not suitable strong for Cryptography. into the control channel of the botnet network which tracks that In the botnet architecture, authenticity and integrity of the bot what kinds of attack has occurred. commands is important. 2.2. Motivation for botnet forensics Analysis 3.3. Cross-analysis of botnet victims The defensive approaches of botnet forensics follows that Balzarotti et al. [10] provide an in depth analysis passive (depend prevention is better than attack, detection and response perspectives. on human action or other) and active(bot frequently use network Botnet forensics ensures that attacker spends more time and energy scanning techniques to get susceptible hosts for spreading infection) to cover his tracks for making the attack. Hackers/ criminals will be measurement study that how data get infected through major more careful to avoid prosecution for their prohibited events. This botnets like Conficker, MegaD, and Srizbi. They observed restriction reduces network crime rate and thus security improves. commonlymalware infected networks. IP address space and Internet Service Providers (ISPs) are also being made answerable physical location of Conficker botnet can be observed by CAIDA. for what’s going on their network. Now a days companies doing A Cross-botnet prediction technique is proposed to predict business on Internet can not hide a security breach and are expected unknown victims of one botnet from the information of the other to prove the state of their security as a compliance measure for botnet if they have similar infection vectors. They will further regulatory purposes. provide new approaches to explain relationships between geopolitical locations and malware infection. 3. BOTNET FORENSIC ANALYSIS 3.4. Financial botnet analysis Various advanced botnet forensic concepts, designs and analysis Financial botnets particularly aimed at carrying out financial approaches were planned to handle the botnet network fraud, popular threat for banking institutions. This is authenticated environments. In recent years, a behavior-based bots detection tool by one of the biggest savings banks in Spain which helps to fight has been developing fast, which gives a serious malware drive. against financial cybercrime. A financial botnet is identifying, 3.1. IRC traffic analysis analyzing, and mitigating through Financial Botnet Analysis. Mazzariello et al. [7] focused that how a mass of probably strong Banking Trojan is a representative malware which evaluated in hosts can be control from being infected. IRC user behavior model financial institution. This analysis automatically emphasized on the organized in a channel to make difference between normal and detection, imagination and sharing intelligence about financial botnet-related activity. They will concentrate on the problem of botnets [11]. detecting botnets, by introducing network traffic analysis 3.5. Graph-based analysis

6 Meenakshi Thapliyal, Anchit Bijalwan, Neha Garg, et al. A Genericgeneric process model for botnet forensic analysis. Forensic Sci Sem, 2018, 8(1): 5-9.

Nagaraja et al. [12] proposed Graph-based analysis, a powerful confirm accurate alarm. Alert and collection of responses during approach because it lacks protocol semantics or packet statistics attack is accomplished, bots are identified. dependence. It depends on being able to accurately model valid 4.3. Incident Response network growth. A botnet can be detected based on the observation Crime response is convenient on the collected information which that an attacker will increase the number of related graph validates and assesses the event by organization strategy, components due to a rapid growth of edges between suspect authorized and business constraint. Securities against future attacks neighboring nodes. BotGrep is a graph theory based approach to and recovers from the existing damage, preplanning is initiated. At botnet detection and analysis. Nodes of communication graph the same time, the decision whether to continue the investigation represent Internet hosts and edges represent communication and gather more information is also taken. A similar response is to between them. be initiated after the investigation phase where the information obtained may requires certain actions to control and reduce the 4. GENERIC PROCESS MODEL attack. We use a diagrammatic approach (Fig.1) for Botnet forensic 4.4. Collection analysis which is based on previous accessible forensics analysis. The traffic data is assembled from the botnet sensors and most of the facts causing minimum contact to the infected machine. Traffic data rapidly change and it is difficult to create the same trace later. TCPDump, Wireshark, Snort tools assist in collections of traffic data. 4.5. Retention Traces of data and logs are stored on a backup device. The original security traffic data are unaffected for legal requirements. Single copy of the data will be analyzed. TCPDump, Wireshark, Snort tools is applied for retention phases. 4.6. Inspection The whole traces data and specific evidences of the attack are composed of combined data format which can be analyzed. TCPDump, Wireshark, Flow-tools, NfDump, Bro, Snort tools provide a proper way to check each and every fact of the attack which improves the security tools. 4.7. Analysis Previous botnet attack’s pattern helps to classify the particular infection sign and reformed to understand the intention and methodology of the attacker and are classified and correlated. The

Fig 1. Generic Process model of Botnet Forensics Analysis data searching and matching attack patterns can be done by Statistical, soft computing and data mining. The tools which support in analysis of botnet attack are TCPDump, Wireshark, TCPFlow, TCPTrace, Olly Dbg, IDA Pro, NetFlow, TCPXtract, 4.1. Preparation of security tool Snort etc. Botnet forensics is relevant only to environments where bot security tools (sensors) like intrusion detection systems, packet 4.8. Investigation analyzers, firewalls, traffic flow measurement software are installed. Botnet forensics is targeted to define communication pathway The required authorizations and legal warrants are obtained so that between an infected machine and back to the origin point of attack. privacy is not dishonored. The preparation phase ensures the Incident response and prosecution of the attacker is used to identify monitoring tools are in well place. the attacker, the tough measurement of the botnet. IP spoofing and stepping stone attack is still prevalent technique of the attacker to 4.2. Recoganizations of bots hide himself. Various security tools produced a specific security infringe to indicate alert during unauthorized actions and inconsistency. 4.9. Results (Max. Possibility of bot’s identified) Different parameters are used for the determination of presence and The legal systematic format is designed to arrive at the nature of the attack. An immediate validation is done to assess and conclusion. The botnet forensic analysis provides visual conclusion confirm the suspected attack and ignore the false alert. TCPDump, of the attacker methodology and feedback for future investigations Wireshark, PADS, Nepenthes, Snort etc. devices are used to to conduct the deployment and improvement of security products. This generic process model of botnet forensics analysis is used in

7 Meenakshi Thapliyal, Anchit Bijalwan, Neha Garg, et al. A Genericgeneric process model for botnet forensic analysis. Forensic Sci Sem, 2018, 8(1): 5-9. both real-time and post attack scenarios [13]. The first five phases correlation between domain names, for speedier detection. It is (including incident action) handle concurrent botnet traffic. The difficult to applied a more system level logs such as Process/service next four phases are common for real time and post attack executions, memory/CPU utilization, disk reads/writes. It is a scenarios. The post attack investigation begins at the inspection biggest challenges server failure based DNS failures, or failures phase, where a copy of the packet capture (lib pcap format) file is related to the name servers, as a means for detecting botnets which given for investigation. The inspection phase fuses inputs from exhibit double fast flux. various sources and identifies attack indicators. The analysis phase 6. CONCLUSION classifies attack patterns using data mining and statistical approaches. The investigation phase involves trace back and Botnet forensics analysis is helps to capture, detect and trace attribution. Finally bot is captured, if results are not satisfactory for particular bot among piles of bots. It also defines the possible further improvement of the security tools. complicated relationship between different kind of bots and tracing the source of attack with hackers. It provides a systematic pathway 5. RESEARCH CHALLENGES to predict future attacks by using intrusion data of previous botnet 5.1 Collection and Detection attack strategy. The main problem is that different kinds of bots having their REFERENCE different characterization makes it very tough to detect these bots, to define the logical relationship between different bots, arises the 1. M. Bailey, E. Cooke, F. Jahanian, X. Yunjing, and M. Karir, "A Survey of Botnet Technology and Defenses," in Conf. For Homeland new challenges to the accuracy of population counting techniques. Security, (CATCH '09. Cybersecurity Applications & Technology, Botnet Overlapping between botnet counting and detection creates 2009) pp. 299-304. 2. J. Govil, "Examining the criminology of bot zoo," in 2007 6th Int. potential hidden relationships among botnets. Conf. on Information,(Communications & Signal Processing, 2007) 5.2. Botnet Size pp. 1-6. 3. E. S. Pilli, R. C. Joshi, and R. Niyogi, "Network forensic frameworks: Each and every minute new advanced bots (TDL-4, Grum) are Survey and research challenges,"(Digital Investigation, 2010) vol. 7, generated in the internet field and its very tough to guess the kind pp. 14-27. 4. S. r. S. C. Silva, R. M. P. Silva, R. C. G. Pinto, and R. M. Salles, of next bot’s attack and how powerful it will be. Botnet forensic "Botnets: A survey," Computer Networks, 2012. analysis involves capturing of bots and tells, what is the position of 5. L. Chaoge, L. Weiqing, Z. Zhiqi, L. Peng, and C. Xiang, "A attack with its maximum possible botnet size. Waledac Botnet size recoverable hybrid C&C botnet," in Malicious and Unwanted Software(MALWARE), 6th Int. Conf. on, 2011, pp. 110-118. estimation is difficult because botnet itself was capable of sending 6. M. A. R. J. Z. Fabian and M. A. Terzis, "A multifaceted approach to about 1.5 billion spam messages a day, or about 1% of the total understanding the botnet phenomenon," in Proc. of the ACM SIGCOMM Internet Measurement Conference (IMC), 2006. global spam volume. It will be difficult to operate on a new botnet 7. C. Mazzariello, "IRC traffic analysis for botnet detection," in especially on the encrypted botnet. Information Assurance and Security, ISIAS'08. Fourth Int. Conf. on, 2008, pp. 318-323. 5.3. Botnet traffic filter 8. Y. Kugisaki, Y. Kasahara, Y. Hori, and K. Sakurai, "Bot Detection It does not contribute to any information between Failover pairs. Based on Traffic Analysis," in Intelligent Pervasive Computing, IPC, Int Conf. on, 2007, pp. 303-306. Failovers or Reboots requires redownload of the Dynamic Database. 9. R. Borgaonkar, "An Analysis of the Asprox Botnet," in Emerging Currently there is no support for IPV6. Security Information Systems and Technologies (SECURWARE), Fourth Int. Conf, 2010, pp. 148-153. 5.4. Investigation 10. R. Sommer, D. Balzarotti, G. Maier, S. Shin, R. Lin, and G. Gu, It is biggest challenge to investigate a robust botnet unwanted "Cross-Analysis of Botnet Victims: New Insights and Implications," in Recent Advances in Intrusion Detection, vol. 6961(Springer Berlin traffic detection algorithms and how to filter botnet command and Heidelberg, 2011) pp. 242-261. control (C&C) traffic early. 11. M. Riccardi, D. Oro, J. Luna, M. Cremonini, and M. Vilanova, "A framework for financial botnet analysis," in Crime Researchers 5.5. Analysis Summit (eCrime), 2010, pp. 1-7. Active analysis is required because honeypot setup is difficult on 12. S. Nagaraja, P. Mittal, C.-Y. Hong, M. Caesar, and N. Borisov, large scale network. Passive anomaly analysis usually independent "BotGrep: finding P2P bots with structured graph analysis," in Proc. of the 19th USENIX conf. on Security, 2010, pp. 7-7. of the traffic content and has the potential to find different types of 13. S. Krasser, G. Conti, J. Grizzard, J. Gribschaw, and H. Owen, botnets (e.g., HTTP, IRC and P2P). "Real-time and forensic network data analysis using animated and coordinated visualization," in Information Assurance Workshop, IAW 5.6. Temporal correlation technique '05. Proc. from the Sixth Annual IEEE SMC, 2005, pp. 42-49. This technique utilize between DNS queries and entropy based

PO STSCRIPT First reported at Conference on Advances in Communication and Control Systems 2013 (CAC2S 2013). CONTACT AUTHORS Meenakshi Thapliyal Department of Computer Science and Engineering, Graphic Era University

8 Meenakshi Thapliyal, Anchit Bijalwan, Neha Garg, et al. A Genericgeneric process model for botnet forensic analysis. Forensic Sci Sem, 2018, 8(1): 5-9.

Dehradun, Uttarakhand, India [email protected] Anchit Bijalwan Department of Computer Science and Engineering, Uttarakhand Technical University Dehradun, Uttarakhand, India [email protected] Neha Garg Department of Computer Science and Engineering, Graphic Era University Dehradun, Uttarakhand, India [email protected] Emmanuel Shubhakar Pilli Department of Computer Science and Engineering, Graphic Era University Dehradun, Uttarakhand, India [email protected]

9 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18.

Peer Reviewed Forensic Science Seminar Received 1 August 2017 ISSN 2157-118X Received in revised form 3 October 2017 Volume 8 Number 1 Accepted 4 October 2018 3 April 2018 Available online 3 April 2018

orensic Accounting on Corporate Governance F Maturity mediated by Internal Audit: A Conceptual Overview

Ali Rehman a, *, Fathyah Hashim b

a. Internal Audit Department, A‘Sharqiyah University, 400, Ibra, Oman. b. Graduate School of Business, Universiti Sains Malaysia, 11800 USM, Penang, Malaysia *CORRESPONDING AUTHOR. E-mail: [email protected]

ABSTRACT Identification and development of forensic accounting as an organization's essential intangible asset has been traced as an immense vital essence that can influence corporate governance maturity. This paper attempts to integrate relevant empirical research and literature to extend the intended potentials of forensic accounting on corporate governance maturity particularly in public listed companies in Oman. Moreover, this paper will identify the role of forensic accounting as preventive measure rather than a detective control which is at present widely understood. The paper also proposes that internal audit poses a mediating role between forensic accounting and corporate governance maturity by linking professional theory and agency theory in its relationship. Since the existing literature provides less evidence that attempt to observe the influence of forensic accounting on corporate governance maturity, this paper offers and calls a promising proposition for future research. KEY WORDS Forensic Accounting, Corporate Governance Maturity, Auditors, Fraud, Forensic science

auditors cannot be considered as the most efficient way to detect or 1. INTRODUCTION limit fraud. It is widely understood and as mentioned by many scholars that According to [7] firm and durable corporate governance poor corporate governance leads to fraud, misappropriation of practices are necessary when owners (shareholders) are not assets and dissatisfied shareholders [1]. It is also stated by [2], that responsible for setting strategy or carrying out business activities, no country is invulnerable to fraud. History and recent past is filled for example publicly traded companies. These corporate with events where organizations demonstrated poor corporate governance practices or codes assist preventing fraud and governance and end up either in filing bankruptcy or incorporate fraudulent activities. These codes can be further strengthened by huge losses in their financial statements (Vinita, 2005). utilization of the right anti-fraud expertise i.e., Forensic Accounting. Vinita [3] and Fernando [4] stated that recent frauds are not only Furthermore, [3] stated that aftermath of corporate accounting due to the failure of corporate governance but also demonstrated scandals and the resultant outcry for transparency and reliability in poor accounting practices which were not detected by the reporting created two outcomes. First is "forensic accounting" and conventional auditors. Controlling ownership and minority second is "innovation in corporate governance". shareholders protection is difficult through conventional corporate Bhasin [8] stated that fraud exists in our society for centuries control mechanism, i.e., it requires process enhancement which can long. It is estimated that the typical business loses 5 percent of its be provided by Forensic accounting [5]. Besides, auditors have revenues each year to fraud, this equates to $50,000 for every $1 denied the role for identifying fraud which had created an million in annual revenues [7]. Such a disastrous situation requires expectation gap between shareholders and auditors [6]. This certain additional steps and one of which is acquiring the assistance expectation gap can only be filled with Forensic Accounting. The of Forensic Accounting towards enhancing and strengthening similar concept is defined by The Association of Certified Fraud corporate governance. In accordance with [9] Forensic Accounting Examiner [7], that financial statement auditors' (i.e., external should be seen as the application of financial skills and auditors) are not fraud examiner and also states that, external investigative mindset conducted with the context to resolve

10 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18. unresolved issues, thus complementing corporate governance in a internal audit department. On the other hand, [1] and [18], stated manner which is not covered by the management's regular devised that Forensic Accounting is positioned to explore the redesign of controls. corporate governance because of its skills and knowledge. Here the 1.1. Background of the Study question arises if Forensic Accounting is positioned to redesign People respond to what is inspected, not what is expected [10]. corporate governance then why Forensic Accounting cannot Several mitigating factors should be available within an become a function of governance management? organization for the entire internal deficiencies. Such internal Furthermore, there are no specific standards available for mitigating factor reduces the time delay for solutions, prevents Forensic Accounting like Internal and External Auditing. In the additional value loss and multiplies the number of tools available absence of these standards, Forensic Accounting as profession is for implementation of solutions. It is widely understood that considered as part of audit instead of being considered as a separate corporate governance is an internal process and majorly persuaded standalone body with own standards and practices. Both Institute of and prompted internally. If internal controls are not strengthened by Internal Auditors [19] and American Institute of Certified Public the use of expertise, then corporate governance is bound to collapse, Accountants [20] negated the role of identifying the fraud and corporate governance maturity (CGM) will be impossible to labeled it as management's responsibility and this role of external achieve and above all, it will also raise the ventures for fraud and auditors as mentioned in [20], is acceptable to Securities and embezzlement. Exchange Commission of United States [2, 21]. Gee [11] opined that ever changing time requires the new Further to above, there is no doubt and in accordance with technologies and expertise as the criminals are always one step regulatory standards, it is the management's responsibility to ahead in thinking and executing the crime. In these regards, the identify and report fraud. However, in doing so management conventional way of fighting or eliminating fraud requires change requires expertise; which is Forensic Accounting. Even though as well. This change can be defined as Forensic Accounting which Forensic Accounting is becoming a necessary part of any is also the necessity and should be considered as an essential organization, it is still not visible within the Codes of Corporate requirement for the implementation of good governance and Governance issued by many authorities and regulators including eventually lead towards attainment of CGM. As mentioned by [12] the recent code issued in Oman [22]. and [13], some tools which were utilized in detecting the fraud Furthermore, the researcher is not able to find specific role of were provided and operated by Forensic Accounting which resulted Forensic Accounting towards: in formation of many governance committees e.g., Committee of  Supporting organization’s operations as risk and fraud Sponsoring Organizations of the Treadway Commission (COSO) compliance testing function. and Sarbanes-Oxley Act (SOX).  Achieving CGM as policy developer, risk assessor, and Forensic Accounting requires providing services which are fraud testing function. beyond the general understanding of corporate people. As Since the existing literature provides less evidence that attempt mentioned by Center for Forensic Studies, it is expected from to observe the influence of forensic accounting on corporate Forensic Accounting to offer reliefs to shareholders by assisting the governance maturity, this paper offers and calls a promising conventional auditors. Moreover, injecting Forensic Accounting proposition for future research. techniques into conventional auditing will be effective to prevent The interests of investors and other stakeholders are usually leakages caused by corporate failure, i.e., provide preventive protected by a three-tier security system which comprises of measures to the organization. governance codes, second is organization's management and last 1.2. Research Gap are the controllers such as auditors [23]. These three tiers lead to Several studies are available which define Forensic Accounting one another for the achievement of CGM and on the whole as an institution or person who appears or calls upon after the fact becomes overall process of CGM. If any one of the three tiers are (i.e. when the fraud has actually happened). For instance, [6, 11, not performing effectively and efficiently, then the whole process 14-16] investigated the role of Forensic Accounting as detection could collapse [24]. In accordance with [3], organizations fail to control, litigation expert, and fraud finder. achieve corporate governance and eventually CGM due to There are only a few studies available which stressed that following reasons: Forensic Accounting should be a part of corporate governance.  Lack of well formulated and implemented policies and However, there is no study that demonstrates the relationship of standards of achieving CGM. Forensic Accounting as a governance management system and as  Lack of honesty, integrity and transparency in financial part of organization's management team. In accordance with [17] reporting governance management are those which are directly responsible • Inefficient and ineffective system of risk assessment for the governance of organization, i.e., board of directors, audit In accordance with the research gap and problems discussed and risk committee, governance and compliance committee and above, the proposed research questions are: Is there any association

11 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18. between FA and CGM? Is there any relationship between FA and CGM is the first step towards understanding where the internal auditors? Is there any correlation between internal auditors organization stands regarding corporate governance. The big and CGM? Whether internal audit mediates relationship between question here is to identify whether the governance can be Forensic Accounting and CGM? measured or not? As stated by [29], the fact that corporate governance encompasses many different kinds of aspects, both 2. LITERATURE REVIEW AND THEORETICAL ‘hard’ and ‘soft’, indicates that perhaps some of them can be BACKGROUND measured more easily than the others. In accordance with [30] and This section will focus on the existing literature relevant to this [31], since hard aspects coincide largely with aspects of compliance, paper. Review of existing studies will provide strong foundation for there may well be a possibility to measure this compliance, understanding that why FA is necessary, how it can strengthen whereas, so-called ‘soft’ aspects (beyond the numbers), which are CGM and how internal audit can perform as mediator between FA to be situated largely in the ‘control environment’, the question is and CGM? This section will also discuss the theories which support not so easily answered. the issues related to enhancement of CGM through Forensic There are several tools available for the measurement of Accounting and internal audit as mediator. governance maturity provided by [24, 32-34] moreover and as 2.1. Past Studies on CGM mentioned by [31] any tool which best serves the organizational Given the ever-growing concerns of various stakeholders, requirements can be utilized. In governance risk and compliance corporate governance has transformed from just being a buzz word capability module defined by [35] there are eight elements that to a dependable and substantial method to ensure preservation of need to be measured and this includes principles and critical shareholder's value. In this regard, CGM evolved into an industry's success factors. If CGM elements are measured properly and skeptical framework which the key attributes associated with levels deployed effectively, they will result in the following universal of maturity on the continuum to becoming a better governed outcomes: organization [24].  Achieve business objectives. There are certain fundamental requirements which are very  Enhance organizational culture. essential towards CGM. These requirements include legal  Increase stakeholder confidence. compliance, accountability, effectiveness, and efficiency of  Prepare and protect organization. operations and ethical behavior [25]. In accordance with [26], it is  Prevent, detect, and reduce adversity and weakness. the responsibility of senior management and board of directors to  Motivate and inspire desired conduct. act towards achievement of CGM. Board of directors and senior  Improve responsiveness and efficiency. management should act as the focal and central point and also act  Optimize social and economic value. as the custodians of corporate governance. All above outcomes cannot be achieved unless element of fraud Corporate governance itself cannot achieve anything unless is eliminated. For this reason, Forensic Accounting plays a major certain measuring tools are incorporated to assess its performance role in the direction of assessment and achievement of CGM. and its impact on organization. As mentioned by [27] corporate Moreover, and in accordance with [26] internal audit is one of the governance can be tick box compliance or can be properly key role players towards the achievement of CGM. It is adequate to implemented to achieve business goals. To measure the corporate mention that CGM is completely dependable on various governance, CGM models are utilized which defines whether components, rules, and regulations. From these components, organizational governance is heading towards accomplishment of Forensic Accounting and internal audit is one of the crucial and goals or not. In accordance with [24] CGM can assist organizations vital functions [4, 8, 36]. to identify where corporate governance competencies may be Corporate governance is a system, which provides the platform positioned on a maturity range and what are the gaps that exist for organizations to perform in a way which is beneficial to all between actual and planned performance. stakeholders. History is filled with events, where organizations CGM is difficult to achieve where there are complex corporate collapse because of poor or non- availability of corporate relationships, weak regulatory requirements, weak legal and governance. Poor corporate governance is referred to a term where judiciary system and scarce human resource capabilities [4]. But if corporate governance is present but it's not implemented in a it is implemented, controlled and achieved then CGM can ensure manner which can ensure the achievement of organizational goals, developed markets, sound policies and procedures and system of i.e. if it is only utilized as compliance check box. Achievement of reward and accountability. Assurance is required towards CGM in the modern financial and business era is not an extra or implementation and achievement, whereas elimination of fraud is choose-able option. CGM is fundamental and primary tool towards required for proper control and safeguarding organization's repute. raising capital, satisfying shareholders and running business These assurances and eliminations cannot be achieved without IA successfully. CGM is essential to all stakeholders including and assistance of FA [25, 28]. employees, vendors, customers, and bankers. CGM also reduces

12 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18. ambiguity and reassures accountability by encouraging adoption of Similar explanation is defined by [2] that FA can positively impact IA and FA in its system. organizational achievement of goals by eliminating fraud risk and 2.2. Past Studies on Forensic Accounting (FA) strengthening internal controls. Therefore, it can be asserted that FA Forensic Accounting is need of an hour and is also essential plays major role in strengthening corporate governance and pillar for all organizations [21]. FA is mainly recognized for achievement of CGM. detecting financial frauds and for this reason, any study of Forensic Although FA is operating without any regulating standards but Accounting is coarse unless it is combined with fraud and tackling its characteristics and skills makes this activity valued and of fraud [37]. Bhasin [38] opined that, FA came into existence appreciated. Imoniana, Antunes, and Formigoni [44] define skills because of the continuous and rigorous frauds, which are not of FA as comprehensive and enough to impact the achievement of detectable by the common auditing tools, additionally, conventional CGM. Moreover, [1, 18] stated that Forensic Accounting is way of auditing is not enough as frauds are still transpiring, despite positioned to explore the redesign of corporate governance because of the presence of policies, governance codes and governance of its skills and knowledge. bodies. In accordance with [7], the majority of the frauds occur in the FA is defined by many scholars as integration and amalgamation category of asset misappropriation, however, they are the lowest in of auditing, investigative and accounting skills. Eliezer and the monetary loss, whereas financial statement fraud is the lowest Emmanuel [39] explained that specialized area within accounting in frequency, yet it is highest in the median loss. It is worth profession is known as Forensic Accounting. Forensic Accounting mentioning that asset misappropriation is controlled and monitored engagements can be suitable for actual or anticipated quarrels or by internal and external auditors, but for detection and prevention lawsuits. Forensic means "suitable for use in court of law" Mitrić, of financial statement fraud, FA is required. One of the distinctive Stanković, and Lakićević [40] established that segment which deals features of corporate governance is to present true and fair view of with the study of fraud and malfeasance is called Forensic the financial statements and if importance is not given in the area of Accounting and person who performs Forensic Accounting is financial statement fraud, then the survival of corporate governance called forensic accountant. The comprehensive definition is defined can be jeopardized or will be at high risk. by [41]: In accordance with [14, 16, 21, 45], FA is only invited once the "...a multidisciplinary field that encourages both a profession fraud is detected. This paper intends to enhance value of FA as part and an industry, where civil or criminal economic and financial of governance management where fraud can be prevented before it claims, whether business or personal, are contested within occurs and FA can further strengthen the internal audit department established political structures, recognized and accepted social and also audit committee in order to achieve CGM. A similar parameters, and well-defined legal jurisdictions, and informed by concept is mentioned by [2], stating that FA with appropriate the theories, methods and procedures from the field of law, auditing, education, training, and experience can provide assistance to accounting, finance, economics, psychology, sociology, and departments and governance bodies to enhance their functions and criminology" responsibilities towards fraud detection and prevention. It is necessary to explain here that there is a difference between In current business environment fraud testing is becoming a forensic accountants, fraud auditors, and financial auditors. requirement. It is an obligation of organization to provide assurance According to [21] financial auditors are only concerned with the to shareholders and its management that organization's financial compliance with applicable standards and laws; fraud auditors statements and operations are free from fraud and FA is the identify the potential fraud and raise the red flag whereas Forensic paramount option in providing such assurance. Although there are Accountants check every area of the potential fraud, identify the no standards or regulations for FA, but ever-changing business reasons, amount, perpetrator and also suggest the way forward and environment created the need for this profession. lesson learned. FA can be considered as competent specialist to influence IA, FA arises from the cause and effect of fraud and technical error especially in case where auditors denied the fact of identifying made by human [14]. Likewise, and as mentioned by [42] the fraud. FA can be an important aspect of any organization which can outbreak of occupational fraud and financial abuse has increased assist in achieving CGM and develops effective mitigating controls. pressure on the accounting and or auditing professions to find Moreover, FA role is not utilized up to its full potential and is only improved ways of uncovering financial crimes and the answer for utilized when fraud occurs i.e., as detective control instead of such problem is FA [38]. preventive measure. As described by [21] the term forensic accounting refers to the 2.3. Internal Audit as Mediator comprehensive view of fraud investigation. It includes preventive In order to achieve CGM, organizations' boards of directors are measures related to frauds and analyzing anti-fraud controls which asserted with responsibility towards maintaining effectiveness of are present in organization's system. Enofe, Ekpulu, and Ajala [43] internal control systems. In this case, the internal audit plays vital opined that FA could directly influence corporate governance. role in assisting the board to discharge its governance

13 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18. responsibilities [17, 46]. Furthermore, and in accordance with [47], approach and will also complete the control cycle. it is the duty of IA to provide assurance for the effectiveness of 3. UNDERLYING THEORIES organizational operations and governance. In these regards, IA becomes an integral part of corporate governance and also vital tool In accordance with [51] theories are formulated to explain, predict towards the achievement of CGM. and understand facts/ events. There are several theories available As mentioned earlier, IA is governed by standards and for corporate governance; however, there are only a few available frameworks. These standards are defined by Institute of Internal in the field of Forensic Accounting and internal audit which can Auditors (IIA). In accordance with IIA internal audit is: directly attribute to this paper. For the purpose of this paper “…Independent, objective assurance and consulting activity following theories are utilized: designed to add value and improve an organization’s operations. It 1. Agency Theory helps an organization accomplish its objectives by bringing a 2. Theory of Profession systematic, disciplined approach to evaluate and improve the There are possibilities of other theories to be included in future effectiveness of risk management, control, and governance research; however, directly linked theories are those which are processes.” mentioned above. Brief of these theories are mentioned below: Further to above, management is responsible for controlling risk 3.1. Agency Theory to ensure that the objectives are met while the scope of internal Agency theory can be defined as shareholders delegate some audit is based on reviewing risk management and internal controls responsibilities to a team of experts while keeping in mind that they [48]. will perform best for the success of their organizations [52]. It is The major task of IA is to provide assurance [28]. These worth mentioning, that success cannot be achieved without assurances are directed towards governance, risk and internal attaining CGM. Similar concept is explained by [53] that an agent controls. Furthermore, and in accordance with [25, 49], stakeholder or agency is hired by one or more person(s), called the principal(s), expectations related to the roles of IA in assessing and making under a contract and is compensated by the principal to achieve assurance on governance maturity have increased and IA plays desired outcomes for the principal. Because the agent is acting on significant role in supporting, planning, assessing, and enhancing behalf of the principal, the principal gives away decision-making CGM. This makes IA an integral part towards providing assurance authority to the agent. Furthermore, and in accordance with [4, 54] for CGM and clearly demonstrates that IA has direct relationship agency theory is applied to explain dominant role of agents on with CGM. corporate governance. Responsibility for CGM falls on an organization's board of Hired agents are directly responsible for the implementation of directors and its audit committee. IA endorses corporate corporate governance and achievement of CGM. This clearly governance by serving as a resource to the board and the audit demonstrates that corporate governance is entirely dependent upon committee by providing assurance on control testing, advisory, and the agents which are hired by principals. These agents form organization insight [50]. Similarly, and in accordance with [36] IA different committees among themselves to further distribute role in risk management is very crucial but internal audit is either responsibilities such as board of directors, audit committee and risk not ready to pursue or is not proactive in pursuing it. However, committee. In accordance with [55], internal audit directly reports with the right level of skills and experience, IA will be able to carry to audit committee and plays vital role concerning organizations out this role very effectively. governance maturity. Further to above, for the fraud risk, it is opined by [19] that Agency theory defines the relationship between management, internal auditors must have sufficient knowledge to evaluate the shareholders and shareholders' nominated directors. Management is risk of fraud but are not anticipated to have skills and expertise of a responsible for the delivery of results and safeguarding the interest person whose primary responsibility is detecting and investigating of shareholders and this interest can only be protected by achieving fraud. It is worth mentioning that this expertise and knowledge can CGM and utilization of control tools which can be categorized as be provided by FA. Likewise, inclusion of FA into IA will not only internal audit and FA. strengthen the CGM but will also complete the IA's process related The focus of corporate governance is security of interest towards to risk evaluation, control testing, and auditing procedures. stakeholders with specific focus on shareholders. As mentioned by IA is a key element of the corporate governance structure, [52] delegation, as per agency theory, creates problems when contributing to the overall quality of the control environment and managers neglect the concerns of their principals (shareholders) internal controls [17]. In current business environment, IA is and put their self-interests on priority line and start gathering compulsory CGM component and plays important role in private benefits. In accordance with [14] common mind set behind strengthening organizational controls. However major suffering of gathering private benefits is that the agents perform entire task and organizations is due to fraud which is lacking in IA's role. By do not want to provide benefits to shareholders who only invest utilizing FA, fraud risk aspect will be dealt with specialized money and perform no operations. Such kind of behavior requires

14 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18. protection and monitoring which can certainly be provided by FA FA, if utilized to its full potential, can assist organizations in and IA. achievement of CGM [38]. According to [7, 21, 57], fraud risk is 3.2. Theory of Profession the highest among all risks which adversely impacts organization's Forensic Accounting is a profession and many specialists in ability to achieve CGM and in accordance with [2], FA is best other fields are seeking to become professionals in this field [13]. positioned to stamp out the fraud risk. This theory defines how profession should work and what are its As mentioned by [52], agency theory defines the concerned and peculiarities and attributes. The theory of profession can also be interrelated parties in corporate governance, whereas, theory of defined as the authority and repute provided by the general public profession as mentioned by [58] emphasizes on Analytical and associations to the profession for protecting public interest. As comprehension; which assists in understanding the characteristics, mentioned by [56], this theory provides an Analytical attributes and structure of the FA profession. Likewise, theory of comprehension which assists in understanding the characteristics, profession also emphasizes on maintaining highest ethical attributes, and structure of the forensic accounting profession. standards, professionalism and protection of public interest, Above mentioned attributes are utilized to assess the profession therefore, FA can be an agent as mentioned in agency theory and of FA, their skills and competence. Assessment is made relating to can positively influence corporate governance in achieving CGM. providing assurance services for stakeholders specially In accordance with the above, researcher will develop the first organization's management and shareholders. Although there are no proposition: official codes available for Forensic Accounting, however, codes Proposition 1: Forensic Accounting has a positive association utilized by Chartered Accountants and Certified Internal Auditors with corporate governance maturity. can be applicable to profession of FA. Likewise, and as informed B. Forensic Accounting and Internal Audit by [4], that it is essential to employ, train and reward professional As per [19] and for the adoption of theory of profession, internal managers who can be held to high standards of competency, ethics audit itself cannot achieve the purpose of this theory, unless, it is and responsibility and thus assists in achieving and maintaining equipped with FA techniques and practices. In order to prevent CGM. rights of the principals, agency theory and for the purpose of Internal audit is already well versed in their profession and professionalism as mentioned in theory of profession internal audit recognized as major tool in relation to CGM however; if FA is requires integration with FA. Furthermore, and in accordance with included in internal audit profession then internal audit along with [43, 59], Forensic Accounting can influence the decision, working FA will become robust, efficient and effective defense control for and opinion of auditors. These influences can range from the achievement of CGM. establishing need or requirement for fraud testing and investigation, Theories assist in defining conceptual framework and also aiding verification of existence or in-existence of fraud and fraud risk variables of this paper. Agency theory defines the relationship compliance adherence. among principals and agents, additionally, this theory also Based on the above, second proposition would be as follows: emphasizes the protection of principal's rights. This protection can Proposition 2: Forensic Accounting has a positive influence on be achieved by introducing FA and internal audit which in turns auditors' opinion. enhances corporate governance and eventually pave the path for C. Internal Audit and Corporate Governance Maturity achieving CGM. Theory of profession force FA and IA to perform Board of directors and audit committee are the integral part of their task in a manner which can assist in protection of public agency theory [4]. They are also responsible for promoting interest. effective corporate governance system and internal audit function plays a key role in assisting the board and audit committee to 4. PROPOSITION DEVELOPMENT discharge their governance responsibilities for the achievement of After tendering above arguments on FA and IA, it is indeed CGM [46]. interesting to further explore the contributions of FA on CGM of As opined by [52, 59], agency theory defines concerned and public listed companies in Oman. The aim of this proposed interrelated parties in corporate governance and it also defines the research is twofold. It is an ultimate aim that this research should necessity of internal audit as part of governance management. In see whether FA improve CGM and also if IA brings positive addition, and according to the theory of profession internal audit is synergy on the relationship between FA and CGM of public listed required to follow its standards which evidently ensure companies. If the outcome shows a positive and effective role on an achievement of CGM [25]. organizational success, then these companies should seriously start Based on the above, the third proposition would be: considering and adopting a method to identify, manage, measure Proposition 3: Internal audit has a positive relationship with and utilize FA and IA to its optimum. CGM A. Forensic Accounting and Corporate Governance D. Mediating Proposition Maturity Mediator or an intervening variable is a type of variable which

15 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18. describes the relationship between an independent and dependent also suggests that FA and IA are interrelated [19, 43, 59]. As variable [60, 61]. Effective and efficient internal audit is mentioned by [64] agency theory suggests a series of procedures to undeniably crucial to guarantee CGM and reliable financial reconcile interest of shareholders and managers. These procedures reporting that eventually boost the investor's confidence in the include monitoring agent’s behavior by implementing auditing and capital markets [17, 54]. As per the Agency theory, CGM cannot be other governance instruments; thus it can be substantiated that achieved unless all of its constituents are functioning effectively CGM is dependent variable which can be influenced by IA and FA. and efficiently i.e. CGM is dependent on other variables' 6. CONCLUSION performance and success. In line with [62], theory of profession defines that the auditor is The intention of this study will be to explore potentials to examine responsible for identifying fraud and its related components the mediating effect of internal audit on FA and CGM as shown in whereas in accordance with [19] IA requires assistance and proposed conceptual framework. The issue of corporate frauds, expertise of a person who is expert in fraud detection i.e. Forensic fraud detecting methods, and corporate governance failure is Accounting. Thus, it could be essential that internal auditor obtains frequently highlighted in several literature. Most literature found is expertise from Forensic Accounting and provides its opinion with towards addressing the issues on FA skills, FA traits, fraud prudence which enhances the likelihoods of accomplishing CGM. detection techniques, managing corporate governance, and also on Therefore, it could be stated that internal audit plays mediator role FA and CG definitions. However, researcher was not able to find between Forensic Accounting and CGM. particular attempt to investigate FA as preventive measure for In similarity to mediating relationship defined by [2] and stated CGM with IA as mediator hence making this paper's propositions above in proposition 1, 2 and 3, there is positive significant direct unique. relationship between FA and CGM, a positive significant direct link Distinctive role of FA is expected to set immense influence on between FA and IA, and a positive significant direct relationship achieving CGM. Hence, this is an indication that FA must be between IA and CGM. For this reason, this study can affirm that IA deployed in a manner to ensure that its properties are identified and mediates the relationship between FA and CGM. optimized to achieve business goals and improve CGM. FA From the above-mentioned theories, the fourth proposition alongside with IA and with its own innovativeness, creativity, skills, would be: and knowledge provides potential for excellent operational system, Proposition 4: IA plays mediating role in the relationship well controlled policies and eventually would result into a positive between FA and CGM impact on profitability, business sustainability, and integrity. Why investigate mediating role? It is mainly because there is 5. CONCEPTUAL FRAMEWORK great possibility that the relationship between FA and CGM is not The proposed mediation model is illustrated in Figure 1 below. The exclusively direct. Conceptually, Internal Audit is a process by model proposes that the independent variable (FA) influences the which an organization secures assurance that the risk exposures it mediator (IA), which then will influence the dependent variable could face is understood and managed appropriately. Moreover, and (CGM). As opined by [63], IA can be examined through its in the absence of specific standards, it is believed that the responsibility to ensure business transparency, examining and effectiveness of FA and its utilization is driven by the effectiveness monitoring effectiveness of internal control and maintain high of internal audit function. integrity in their corporate business activities and operations. Notably, this paper positions itself in extending previous studies on FA to a new perspective by incorporating its preventive role towards strengthening corporate governance maturity. This is P1 expected to contribute and bring new insights and views on FA which have not been explored before. This should offer vast Corporate opportunities for future researchers to investigate further via Forensic P2 Internal P3 Governance diverse perspective pertaining to FA. Furthermore, this paper Accounting Audit Maturity expects that the proposed theoretical framework developed will be (IV) (Mediator) P4 (DV) extended to an empirical phase following this paper.

REFERENCE Fig. 1. Conceptual Framework 1. Bhasin, M. L, Corporate Governance and Forensic Accountant: an Exploratory Study. Journal of Accounting – Business & Management

Vol. 20, Issue 2, pp. 55-83, 2013. FA is proposed as the independent variable based on the fact of 2. Johnson, P., Ayoib, C.-A. R., and Shamsiah, S, Forensic Accounting literature and previous research, which reveals that FA can have and Fraud: Capability and Competence Requirements in Malaysia. Journal of Modern Accounting and Auditing., Vol. 10 Issue 8. pp strong influence on CGM [1, 18]. Simultaneously, past research 825-834, 2014.

16 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18.

3. Vinita, R. Corporate Governance and the Forensic Accountant. The avoid becoming a ‗tick- box’ exercise and encourage a longer term CPA Journal. Volume 75 Issue 3. pp 68-78, 2005. view of business performance. Mazars, 2015. 4. Fernando, A, Corporate Governance: Principles, Policies and 28. Roth, James, Best Practices: Evaluating the Corporate Culture. Practices. Pearson, 2009. Institute of Internal Auditors Research Foundation (IIARF), 2010. 5. Fan, J. P., and Wong, T. J, Do External Auditors Perform a Corporate 29. Cattrysse, J, Reflections on Corporate Governance and the role of the Governance Role in Emerging Markets? Evidence from East Asia. internal auditor. Roularta Media Group, pp 1-64, 2005. Journal of Accounting and Research. Vol 43 Issue 1. pp 35-72, 2005. 30. Meredith, E., and Robyn, C, Corporate Governance and Performance: 6. Adedire, T. O, Relationship between Forensic Accountants’ An Exploration of the Connection in a Public-Sector Context. Competences and Audit Expectation Gap: Evidence from Nigeria University of Canberra, 2005. Money Deposit Banks. Jomo Kenyatta University of Agriculture and 31. Roberta, R., Sanjai, B., and Brian, B, The Promise and Peril of Technology, Nairobi, Kenya, 2016. Corporate Governance Indices. Yale Law School; Yale Law School 7. ACFE, Report to the Nations on Occupational Fraud and Abuse. Legal Scholarship Repository. Vol 8 Issue 8. pp 1803-1882, 2008. ACFE Magazine, 2016. 32. OECD, Risk Management and Corporate Governance, Corporate 8. Bhasin, M. L, Integrating Corporate Governance and Forensic Governance, OECD Publishing, 2014. Accounting: A Study of an Asian Country– International Journal of 33. Committee of Sponsoring Organizations of the Treadway Management Sciences and Business Research, Vol. 6, Issue 1. pp Commissions (COSO, Enterprise Risk Management: Aligning risk 31-52, 2017. with strategy performance. Committee of Sponsoring Organizations 9. Jafaru Jimoh, Philipa O. Idogho, and Francis O. Iyoha (2012). of the Treadway Commissions, 2016. Corporate Governance and Ethical Sensitivity of Directors of Listed 34. OECG, Red Book GRC Capability Model Version 2.1, OECG Companies: Evidence from Nigeria. Global Journal of Management publishing, 2012. Science and Technology Vol. 1 Issue. 3. pp 20-28 35. Switzer, C. S., Mitchell, S. L., and Mefford, J. L, Governance Risk 10. Munshi, A, Performance Management Based on The Balanced Compliance Capability Model Version 3. OECG. Volume 1 Issue 3, Scorecard, Institute of Internal Auditors, Oman Chapter Conference 2015. Presentation, 2016. 36. Griffiths, P, Risk Based Auditing. Ashgate Publishing Group, 2005. 11. Gee, S, Fraud and Fraud Detection: A Data Analytics Approach. John 37. Leonard, Fraud Testing in Audit Program. Muscat, 2016. Wiley & Sons, Inc., Hoboken, New Jersey, 2014. 38. Bhasin, M. L, Contribution of Forensic Accounting to Corporate 12. Baird, J., and Zelin, R. An examination of the impact of obedience Goverance: An Exprolatory Study of an Asian Country– International pressure on perceptions of fraudulent acts and the likelihood of Business Management 10(4), pp 479-492, 2016. committing occupational fraud. Journal of Forensic Studies in 39. Eliezer, O., and Emmanuel, B, Relevance of Forensic Accounting in Accounting and Business, Statesboro, GA: Center for Forensic the Detection and Prevention of Fraud in Nigeria. International Studies in Accounting and Business, Georgia Southern University, Journal of Accounting Research Vol. 2, No. 7, pp 17-25, 2015. Volume 1, No 1, 1-14, 2009. 40. Mitrić, M., Stanković, A., and Lakićević, A, Forensic Accounting – 13. Bhasin, M. L, Corporate Governance and Forensic Accountants’ Role: the Missing Link in Education and Practice. Management Journal for Global Regulatory Action Scenario – International Journal of Theory and Practice Management. Issue 65. pp 41-56, 2012. Accounting Research Vol. 1, No.1.pp 54-86, 2013. 41. Wm. Dennis, H and James A, D, Research in Forensic Accounting - 14. Adrian, N. K., Lawrence, A., and Cristal Lee, A. S, Forensic What Matters. Journal of Theoretical Accounting Research. Vol 9 Accounting: Public Acceptance towards Occurrence of Fraud Issue 2. pp 40-70, 2014. Detection. International Journal of Business Management, Vol. 4, No 42. Effiok, S. o., and Eton, o. e, Ensuring quality control of forensic 11, pp.145-150, 2009. accounting for efficient and effective corporate management. Global 15. Leo, M., and Stephen, O, Introduction to Forensic Accounting – The journal of social sciences vol 12. pp 27-33, 2013. Story behind the Numbers. Financial Executive International – RSM, 43. Enofe, A., Ekpulu, G., and Ajala, T. O, Forensic Accounting and 2015. Corporate Crime Mitigation. European Scientific Journal edition 16. Nigrini, M. j, Benford‘s Law Applications for Forensic Accounting, vol.11, No.7, 2015. Auditing, and Fraud Detection. John Wiley & Sons, Inc., Hoboken, 44. Imoniana, J. O., Antunes, M. T., and Formigoni, H, The forensic New Jersey, 2012. accounting and corporate fraud. JISTEM J.Inf.Syst. Technol. Manag. 17. Hermanson, H. M., and Rama, D. B, Pressure on Internal Auditors to vol.10 Issue1. pp 119-144, 2013. Alter Findings. Journal of Forensic & Investigative Accounting. Vol 8 45. Charles, D., Ramona, F., and Suzanne, O, Characteristics and Skills Issue 2. pp 241-256, 2016. of the Forensic Accountant. American Institute of Certified Public 18. Ali, M., and Oseni, Corporate Governance, Transparency And Accountants, 2009. Shareholders’ Rights: The Role Of The Forensic Accountants In 46. Florea, R., and Florea, R, Internal Audit and Corporate Governance. Propelling Economic Growth. Multidisciplinary Journal Economy Transdisciplinarity Cognition. Vol 16 Issue 1. pp 79-83, of Research Development, Volume 15, No 1, 2010. 2013. 19. IIA, I. o,. Internal Auditing and Fraud. The Institute of Internal 47. International Organization of Supreme Audit Institutions, The Auditors-IPPF Practice Guide, 2016. International Standards of Supreme Audit Institutions; The Auditor’s 20. American Institute of Certified Public Accountants - AICPA, Responsibilities Relating to Fraud in an Audit of Financial Generally Accepted Auditing Standards GAAS. AICPA, 2013. Statements. International Organization of Supreme Audit Institutions, 21. Singleton, T., and Singleton, A, Fraud Auditing and Forensic 2016. Accounting 4th Edition. Chicago: John Wiley & Sons, 2010. 48. Pickett, K. H, The Essential Guide to Internal Auditing, 2nd Edition. 22. Capital Market Authority-CMA, Oman Code of Corporate A John Wiley & Sons, Ltd Publication, 2011. Governance for Public Listed Companies. Capital Market Authority, 49. IFA, I. F, International standard on auditing 240: the auditor’s 2016. responsibilities relating to fraud in an audit of financial statements. 23. Bhasin, M. L, Communion of Corporate Governance and Forensic International Federation of Accountants, 2009. Accounting: A Study of an Asian Country– British Journal of 50. Mahd Ali Al- Jabali, Osama Abdalmanam, Khalifeh N. Ziadat, Research, 10 (4), pp 143-162, 2016. Internal Audit and its Role in Corporate Governance. Middle Eastern 24. Deloitte, T, The centre for corporate governance: Board Leadership. Finance and Economics. Issue 11. pp 161-176, 2011. Deloitte & Touche LLP Internal Audit Professionals, 2010. 51. Swanson, R. A., and Chermack, T. J,. Theory Building in Applied 25. IIA Research Foundation, Improving Organizational Governance Disciplines. Berrett-Koehler Publishers, 2013. through Implementing Internal Audit Standard 2110. IIA Research 52. Afza, T., and Nazir, M. S, Theoretical Perspective of Corporate Foundation,2011. Governance: A Review. European Journal of Scientific Research, Vol. 26. Wilkinson, N., and Plant, K, A framework for the development of an 119, No 2, pp. 255-264, 2014. organizational governance maturity model: a tool for internal auditors. 53. Miles, J. A, Management and Organization Theory. Wiley, 2012. Southern African Journal of Accountability and Auditing Research 54. Earnest, D. F., and Sofian, S, The Mediating Role of Corporate Vol 13 Issue 1. pp 19-31, 2012. Governance on Intellectual Capital and Corporate Performance. 27. Mazars, EcoDa, and Ecgcn, Corporate Governance codes should Journal of Economics, Business and Management, Vol. 1, No. 4,pp

17 Ali Rehman, Fathyah Hashim. Forensic accounting on corporate governance maturity mediated by internal audit: A conceptual overview. Forensic Sci Sem, 2018, 8(1): 10-18.

339-343, 2013. Wiley & Sons, Inc, 2003. 55. IIA Research Foundation, I, Internal Audit reporting relationships: 61. Wu, A. D., and Zumbo, B. D, Understanding and Using Mediators serving two masters. IIA Research Foundation, IIARF, 2003. and Moderators. Social Indic Research, vol. 8, no. 7. pp 367-392, 56. Chen, Y., and Akkeren, J. V, The Theory of Profession: 2008. Accountability, qualifications, entry and ethics - a preliminary 62. Sijpesteijn, F. K, The value relevance of auditors’ communications discussion and early findings on the current state of forensic financial statement users’ understanding of the messages in the audit accountancy in Australia, 2013. report. Erasmus Universiteit Rotterdam, 2011. 57. Eyisi, A. S., and Ezuwore, C. N, The Impact of Forensic Auditors in 63. Alic, M., and Rusjan, B, Contribution of the ISO 9001 internal audit Corporate Governance. Research Journal of Finance and Accounting, to business performance. International Journal of Quality & Vol.5, No.8. pp 31-39, 2014. Reliability Management, Volume 27 Issue: 8, pp. 916-937, 2010. 58. Candilis, P, The Revolution in Forensic Ethics: Narrative, 64. Roslan, A., and Dahan, H. M, Mediating Effect of Enterprise Risk Compassion, and a Robust Professionalism. Psychiatric Clinics of Management on Internal Audit and Organizational Performance: A North America. Vol 32 Issue 2. pp 423-435, 2009. Conceptual Framework. International Journal of Commerce, 59. Micheal. B. Adams, ―Agency Theory and the Internal Audit‖, Business and Management (IJCBM). Volume 2 Issue 4. pp 212- 215, Managerial Auditing Journal, Vol. 9 Issue: 8, pp.8-12,1994. 2013. 60. Sekaran, U, Research Methods for Business: A Skill-Building. John

POSTSCRIPT First reported on the 1st Economics and Business International Conference 2017 (EBIC 2017) CONTACT AUTHORS Ali Rehman Internal Audit Department, A‘Sharqiyah University, 400, Ibra, Oman [email protected] Fathyah Hashim Graduate School of Business, Universiti Sains Malaysia, 11800 USM, Penang, Malaysia [email protected]

18 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

Peer Reviewed Forensic Science Seminar Received 2 April 2018 ISSN 2157-118X Received in revised form 2 April 2018 Volume 8 Number 1 Accepted 2 April 2018 3 April 2017 Available online 3 April 2018

wo Different Approaches to Handle Landmark T Location Uncertainty in Skull-face Overlay: Coevolution vs Fuzzy Landmarks

Óscar Ibáñez a, *, Óscar Cordón a, b, Sergio Damas a

a. European Centre for Soft Computing, C/ Gonzalo Gutierrez Quirós, s/n, 33600 Mieres, Asturias, Spain. b. Department of Computer Science and Artificial Intelligent (DECSAI and CITIC-UGR), University of Granada, C/ Daniel Saucedo Aranda, s/n, 18071 Granada, Spain. *CORRESPONDING AUTHOR. E-mail: [email protected]

ABSTRACT Craniofacial superimposition is a forensic process where photographs or video shots of a missing person are compared with the skull that is found. By projecting both photographs on top of each other (or, even better, matching a scanned three-dimensional skull model against the face photo/video shot), the forensic anthropologist can try to establish whether that is the same person. The whole process is influenced by inherent uncertainty mainly because two objects of different nature (a skull and a face) are involved. In this paper, we propose a novel approach, a cooperative coevolutionary algorithm, to deal with the use of imprecise cephalometric landmarks in the skull-face overlay process. Following this approach, we are able to look for both the best transformation parameters and the best landmark locations at the same time. Coevolutionary skull-face overlay results are compared with our previous fuzzy-evolutionary automatic method over six skull-face overlay problem instances corresponding to three real-world cases solved by the Physical Anthropology lab at the University of Granada (Spain). Promising results have been achieved though the robustness of the method should be improved. KEY WORDS Forensic identification, Craniofacial superimposition, Skull-face overlay, Coevolution, Cooperative coevolutionary algorithm, Fuzzy landmarks, Evolutionary algorithms, Forensic science

CMA-ES demonstrated to be the most accurate and robust 1. INTRODUCTION approach. Craniofacial superimposition (CS) [1, 2, 3] is a forensic process The SFO process is influenced by inherent uncertainty since two where photographs or video shots of a missing person are objects of different nature are involved (a skull and a face) [6]. In [7] compared with the skull that is found. By projecting both we studied in detail the sources of uncertainty related to the SFO photographs on top of each other (or, even better, matching a task and proposed the use of imprecise landmarks to overcome scanned three-dimensional skull model against the face most of the limitations associated with them. Using imprecise photo/series of video shots), the forensic anthropologist can try to landmarks, forensic anthropologists are able to properly deal with establish whether that is the same person. This skull-face overlay two different problems. On the one hand, the difficult task of (SFO) process is usually done by bringing to match some locating anthropometric landmarks invariantly. On the other hand, corresponding anthropometrical landmarks on the skull and the the identification of a big enough set of non-coplanar landmarks. face. These imprecise landmarks were modeled following the fuzzy set SFO is known to be one of the most time-consuming tasks for theory-based approach and their use led to a really important the forensic experts [4] (it takes up to 24 hours in many real-world performance improvement of our automatic method. situations). However, the fuzzy-evolutionary approach relies on more As it is explained in [5], we have developed an automatic method computational operations due to the fact that distances between a that overlays a 3D model of the skull over a 2D photograph of the crisp point and a fuzzy set of points must be computed. Hence, the face of the missing person. With that aim, we studied and run time required by the algorithm increased. In particular, the crisp experimented with different evolutionary approaches in [5], where landmark approach ranged in 10-20 seconds per run while the

19 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

fuzzy landmark approach ranged in 2-4 minutes. 11 1 With the aim of decreasing the run time needed without losing 11 1 = = accuracy, we propose a new evolutionary approach based on a ⋮ ⋮⋮⋮ ⋮ ⋮⋮⋮ [8] coevolutionary algorithm (CEA) . It also takes advantage of 11 1 imprecise landmarks but, unlike the fuzzy approach where

distances between a fuzzy set and a crisp point have to be the overlay procedure aims to solve a system of equations with calculated (computationally costly), it only needs to calculate the following twelve unknowns: a rotation represented by an axis Euclidean distances between pairs of crisp points. (dx, dy, dz) and angle (θ), a center of mass (rx, ry, rz), a translation The novel proposal is tested on six SFO problem instances vector (tx, ty, tz), a uniform scaling (s), and a 3D-2D projection derived from three real-world identification cases solved by the function given by a field of view (ϕ). The corresponding equation Physical Anthropology Lab at the University of Granada (Spain). system is as follows: The structure of the paper is as follows. In Section 2 we describe

the SFO problem and our previous evolutionary method to deal −1 −1 −1 F = f(C) = C · (A · D1 · D2 · Rθ · D2 · D1 · A ) · S · T · P with it. Then, we study the sources of uncertainty associated with (1) the SFO task (Section 3) and summarize our previous fuzzy

approach to tackle them. Section 4 is devoted to introduce our −1 −1 −1 where R = (A · D1 · D2 · Rθ · D2 · D1 · A ) represents a rotation proposal based on a cooperative coevolutionary algorithm. In matrix to orient the skull in the same pose of the photograph. S, T, Section 5, we test and compare the new proposal against the and P are uniform scaling, translation, and perspective projection fuzzy-evolutionary one over six problem instances. Finally, we matrices, respectively. The interested reader is referred to [10] for a present some concluding remarks and future works in Section 6. detailed description of the matrices in Equation 1 and their relation 2. SKULL-FACE OVERLAY IN CRANIOFACIAL with the twelve unknowns of the problem, as well as to [5] for a SUPERIMPOSITION deeper explanation. Different definitions of the fitness function were studied and the The success of the SFO process requires positioning the skull in the one achieving the best results was the mean error (ME)1: same pose of the face as seen in the given photograph (provided by the relatives of the missing/deceased person). The orientation ∑ − process is a very challenging and time-consuming part of the CS = (2) technique [4].

Most of the existing SFO methods are guided by a number of where || · || is the 2D Euclidean distance, N is the number of anthropometrical landmarks located in both the skull and the considered landmarks (provided by the forensic experts), cli photograph of the missing person. The selected landmarks are corresponds to every 3D craniometric landmark, fli refers to every placed in those parts where the thickness of the soft tissue is low. 2D facial landmark, f is the function which defines the geometric The goal is to ease their location when the anthropologist must deal 3D-2D projective transformation, and f(cli) represents the projected with changes in age, weight, and facial expressions. skull 3D landmark cli in the image/photo plane. Once these landmarks are available, the SFO procedure is based Solving the SFO problem in the latter fashion results in a really on searching for the skull orientation leading to the best matching complex optimization task, with a highly multimodal landscape, of the two sets of landmarks. and forensic experts demand very robust and accurate results. This In view of the task to be performed, the relation of the desired complex landscape leaded us to tackle the problem considering procedure with the image registration (IR) problem in computer robust evolutionary algorithms (EAs) [11] to search for the optimal vision [9] can be clearly identified. Following a 3D-2D IR approach values of the twelve registration transformation parameters. In [5], we aim to properly align the 3D skull model and the 2D face CMA-ES [12] and different real-coded genetic algorithms [13] were photograph in a common coordinate frame. The required applied, achieving very promising results in some problem perspective transformation to be applied on the skull was modeled instances. in [5] as a set of geometric operations involving twelve parameters/unknowns which are encoded in a real-coded vector to 3. SOURCES OF UNCERTAINTY IN SKULL-FACE represent a superimposition solution. OVERLAY Formally, SFO can be formulated as follows: given two sets of 2D facial and 3D cranial landmarks, F and C, respectively, both The whole CS process is influenced by uncertainty. In particular, comprising N landmarks: SFO is affected by two different sources of uncertainty of different

1 Notice that, mean square error is not used because of its negative effect when image ranges are normalized in [0, 1].

20 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

nature. On the one hand, there is an inherent uncertainty associated unsuitable spatial distribution in the quality of the SFO results. This with the two different kinds of objects involved in the process, i.e. effect happens when the landmarks guiding the optimization a skull and a face. On the other hand, there is also uncertainty method are coplanar or near-coplanar. associated with the 3D-2D overlay process that tries to Due to landmark coplanarity the Equation 1 becomes superimpose a 3D model over a 2D image. Following, we will undetermined (or near-undetermined), i.e. there is uncertainty summarize the main characteristics of both sources and we will (there is not enough information or it is imprecise) regarding which review our previous proposal based on the use of fuzzy landmarks of the possible solutions is the best. As a consequence, it is not to overcome most of the limitations associated with the latter possible to numerically distinguish among the different resulting sources of uncertainty. (after a search process) sets of projection parameters which 3.1. Uncertainty inherently associated with the objects under originate different SFO results. In [15] we experimentally study demonstrated the strong negative impact of coplanar landmarks in We have identified two inherent sources of uncertainty regarding the quality of the SFO results derived by our automatic procedure. the handled objects (a skull and a face) and their relationship. On Having a reasonable number of anthropometrical landmarks the one hand, the landmark location uncertainty is related to the located in different planes becomes a real need for the good extremely difficult task of locating the points in an invariable place, performance of the method. since the definition of any anthropometric landmark is imprecise in 3.3. Imprecise landmarks: a fuzzy set-based approach its own. Indeed, every forensic anthropologist is prone to locate the Our approach is based on allowing the forensic anthropologist to landmarks in a slightly different position [14], regardless the means perform an imprecise location of cephalometric landmarks. By used to represent the involved objects (skull and face), i.e. 3D using imprecise landmarks, (s)he can locate the landmark as a model, 2D photo, video shot, . . . region instead of as a crisp point as usual. The size of the region On the other hand, the landmark matching uncertainty refers to defined by the forensic expert will become a measure of the the imprecision involved in the matching of two sets of landmarks landmark uncertainty (the broader the region, the higher the corresponding to two different objects: a face and a skull. There is a uncertainty in the location of that landmark). clear partial matching situation. The correspondence between facial Notice that, by marking landmarks in an imprecise way, we and cranial anthropometric landmarks is not always symmetrical manage to solve the problems related to three of the four and perpendicular, some landmarks are located in a higher position uncertainty sources analyzed at the same time. First, the inherent in the alive person face than in the skull, and some others have not uncertainty of the landmark location in the missing person got a directly related landmark in the other set [2]. Besides, the photograph can be properly tackled. In the same way, the forensic facial soft tissue depth varies for each cephalometric landmark and experts are able to deal with the location of landmarks whose also for every person group (regarding age, race, and sex). position they cannot determine accurately due to the photograph 3.2. Uncertainty associated with the 3D skull model-2D face conditions with the proper level of confidence (using imprecise photo overlay process regions of different sizes). As a consequence, we will allow them to deal with the extremely difficult task to increase the number of The uncertainty associated with the 3D skull-2D face overlay is selected landmarks. As explained, those additional landmarks are not inherent to the objects themselves but to our approach, where essential to face the coplanarity problem in the automatic search of we try to overlay a 3D model and a 2D image. As done in Section the best SFO. Only the landmark matching uncertainty has been 3.1, we can also distinguish between landmark matching and left for future works. landmark location uncertainty. However, the nature of these two The imprecise landmark location approach was implemented sources of uncertainty is different in the current case. through two alternative models of imprecise landmarks: weighted On the one hand, the landmark location uncertainty refers to the and fuzzy ones. We summarize the fuzzy one since it achieved difficulty of locating landmarks with the accuracy required for the better skull-face overlay results. For a detailed description of both automatic overlay of a 3D skull model and a 2D face photo. The approaches the interested reader is referred to [7]. ambiguity may arise from reasons like variation in shade The fuzzy landmark approach is based on allowing the forensic distribution depending on light condition during photographing, experts to locate the cephalometric landmarks using ellipses and on unsuitable camera focusing, poor image quality, face pose in the considering fuzzy sets to model the uncertainty related to them. photograph, partial or whole landmark occlusion, etc. Besides, we have also considered fuzzy distances to model the Forensic experts are prone to only locate those landmarks which distance between each pair of craniometric and cephalometric can be unquestionably identified in the skull and in the face. landmarks. Therefore, they are usually only able to locate a reduced set of all We have defined a fuzzy landmark as a fuzzy convex set of the available cephalometric landmarks. points having a nonempty core and a bounded support. That is, all On the other hand, the matching uncertainty refers to the its α-levels are nonempty bounded and convex sets. In our case, negative influence of a small number of landmarks with an

21 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

since we are dealing with 2D photographs with an x × y resolution, Therefore, we modified the original definition of our we have defined the fuzzy landmarks as 2D masks represented as a evolutionary SFO technique’s fitness function as follows:

matrix M with mx × my points (i.e., a discrete fuzzy set of pixels). Each fuzzy landmark will have a different size depending on the ∑ ∗, = (3) imprecision on its localization but at least one pixel (i.e. crisp point related to a matrix cell) will have membership with degree one.

These masks are easily built starting from two triangular fuzzy where N is the number of considered landmarks; cli corresponds sets and modeling the approximate vertical and horizontal to every 3D craniometric landmark; f is the function which defines position of the ellipse representing the location of the landmark, the geometric 3D-2D transformation; f(cli) represents the position thus becoming two-dimensional fuzzy sets. of the transformed skull 3D landmark cli in the projection plane, An example of these fuzzy cephalometric landmarks is given in that is to say, a crisp point; i represents the fuzzy set of points of Figure 1, where the corresponding membership values (calculated i each 2D cephalometric landmark; and, finally, d∗(f(Ci), ) is the using the product t-norm) of the pixels of one of those landmarks is distance between a point and a fuzzy set of points. depicted on the right. 4. A COOPERATIVE COEVOLUTIONARY ALGORITHM TACKLING THE LANDMARK LOCATION UNCERTAINTY IN CRANIOFACIAL SUPERIMPOSITION

EAs have been applied to many types of problem domains, such as parameter optimization and machine learning. A very natural, and increasingly popular extension when problems domains are potentially complex, or when it is difficult or impossible to assess an objective fitness measure for the problem, is the class of so-called coevolutionary algorithms (CEAs) [8]. In such algorithms, fitness itself becomes a measurement of interacting individuals. This ostensibly allows the potential for evolving greater complexity by allowing pieces of a problem to evolve in tandem, as well as the potential for evolving solutions to problems in which such a subjective fitness may, in fact, be necessary (i.e., game playing

strategies). Hence, a coevolutionary algorithm is an EA (or Fig. 1. Example of fuzzy location of cephalometric landmarks (on the collection of EAs) in which the fitness of an individual depends on left) and representation of an imprecise landmark using fuzzy sets (on the relationship between that individual and other individuals [18]. the right). Depending on the nature of the interactions among individuals

we can distinguish between competitive and cooperative CEAs. In

the former, each specie competes with the rest. In the latter, named Now we can calculate the distance between a point (which will cooperative coevolutionary algorithms (CCEAs) [19], all the species be the pixel constituting the projection of a 3D craniometric collaborate to build a solution for the problem. The landmark on the 2D face photo) and a fuzzy landmark (the discrete originally-stated aim of CCEAs was to attack the problem of fuzzy set of pixels representing the imprecise position of the evolving complicated objects by explicitly breaking them into parts, cephalometric landmark). evolving the parts separately, and then assembling the parts into a If we denote as di = d(x, ) the distance from point x to the working whole. α-level set (in this specific case, this is the minimum This is the goal of the current proposal in which two different but Euclidean distance from point x to the all the points in αi), then the complementary problems arise. On the one hand, we want to know distance from the point to the fuzzy landmark , can be expressed the best set of transformation parameters resulting in the best by: possible SFO. On the other hand, SFO quality is measured based

on the distances among two set of landmarks, where the location of ∑ ∙ ∗, = the cephalometric one is uncertain. We only know that they are ∑ located somewhere inside a region delimited by the forensic expert

(imprecise landmark). Hence, we can try to find the exact locations Note that the implemented distance between a point and a fuzzy of the cephalometric landmarks in that region. However, the only set of points is quite similar to that defined in [16]. In fact, it has way to determine them is by looking for the best SFO, assuming been already used in other image processing applications in [17]. that the most precise location of the landmarks implies the chance

22 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

to achieve the most precise SFO. That is to say, we need to look, at In the following section we show the set of parameters that the same time, for the best set of transformation parameters and for results in the best SFO results. the most precise cephalometric landmarks locations. 5. EXPERIMENTS To do that, we have implemented a CCEA where two populations optimize the set of transformation parameters and the Our experimental study will involve six different SFO problem location of the cephalometric landmark, respectively. For the instances corresponding to three real-world cases previously former population we consider a coding scheme representing the addressed by the staff of the Physical Anthropology lab at the twelve registration transformation parameters in a vector of real University of Granada (Spain) in collaboration with the Spanish numbers while for the latter we consider a variable-length scientific police. integer-valued chromosome, whose size corresponds with the All those identification cases were positively solved following a number of pair of landmarks used. Both populations need to computer-supported but manual approach for SFO. We will collaborate/interact to construct a solution for the main problem, consider the available 2D photographs of the missing people and that is to achieve the best possible SFO. their respective 3D skull models acquired at the lab by using its The question of how a practitioner determines collaborators or Konica-Minolta 3D © Lasserscanner VI-910. competitors may be among the most important design decisions for The experiments developed in this section are devoted to study the successful application of CEAs [20]. The most obvious (and the performance of the coevolutionary-based approach to model the computationally expensive) method to evaluate an individual in a imprecise location of cephalometric landmarks within our SFO coevolutionary setting is for an individual to interact with all method in comparison with the previous fuzzy set-based proposal. possible collaborators or competitors. In the case of binary With this aim, we first show the parameter setting considered. Then, interactions, this is sometimes called complete pairwise interaction. we introduce each of the six selected real-world SFO problems to An alternative extreme is for an individual to be involved in only a be tackled together with the obtained results and their analysis. single interaction. Such a choice leaves open the obvious question 5.1. Experimental design of how to pick the collaborator or competitor. Of course, between We used CMA-ES as the evolutionary algorithm for the case of these two extremes there is a whole host of possibilities that fuzzy set-based approach [7] guided by the corresponding objective involve some subset of interactions. Again, function (Equation 3) and we considered the following set of collaborators/competitors for such interactions may be chosen in a parameters: variety of ways from uniformly random, to fitness biased methods. There are mainly three attributes of this choice, suggesting a wide initial (mutation distribution variance) = 0.1 range of possible strategies. They are briefly described below: (population size, offspring number) = 100  Interaction sample size: the number of (number of parents/points for recombination) = 15 collaborators/competitors from each population to use for a given fitness evaluation. The rest of the parameters are the default ones, reported in [12].  Interaction selective bias: the degree of bias of choosing a For the case of the coevolutionary approach, we performed many collaborator/ competitor. For example, an individual could different combinations of parameters 2 under the restriction of be selected either randomly or based on the fitness value. 560,000 evaluations of the fitness function (according to the  Interaction credit assignment: the method of credit CMA-ES conditions). The set of parameters considered are the assignment of a single fitness value from multiple following: interaction-driven objective function results. The main examples here are the minimum, the maximum and the # generations = 600 mean of all the fitness values. T size = 450 In our implementation of the CCEA we tested many different L size = 25 options for these three parameters. As interaction selective bias we T and L crossover probability = 0.9 use both the best individuals and random ones. The minimum, the T and L mutation probability = 0.2 maximum and the mean were tested as credit assignment T and L tournament size (for RCGA) = 2 mechanisms. SBX η parameter (for population T) = 0.1 The transformation parameters population uses SBX crossover Interaction sample size and selective bias: [5] while the landmark locations population uses multi-point # best selected individuals in T = 1 crossover. Random mutation is used in both cases, constraining the # random selected individuals in T = 1 possible values for the landmark location to all the pixels inside the # best selected individuals in L = 2 region corresponding to the imprecise landmark the forensic expert located in the image. 2 These preliminary experiments are not included because of a lack of space

23 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

# random selected individuals in L = 2 approach (first photograph of Figure 3). Interaction credit assignment = m

where T is the population of the transformation parameters, L is the population of landmark locations and m corresponds to the minimum value from the multiple interaction-driven objective function results. Thirty independent runs were performed for each case in order to avoid bias in the results and to compare the robustness of both proposals. We would like to have a quantitative measure allowing us to benchmark the achieved outcomes. Unfortunately, the ME values obtained by each approach are not fully significant to perform a comparison because of the different objective functions to be minimized. In addition, there is not a direct correspondence between ME values and the visual representations as it was pointed out by the experts in [7]. Fig. 3. Case studies: photograph of the missing person with the Due to the latter reasons, we adopted an alternative, specifically corresponding set of imprecise landmarks. designed image processing scheme to evaluate the performance of First row, first image corresponds to case study 1. First row, second image every SFO approach called “area deviation error” (ADE) [7, 15]. In corresponds to case study 2. The other four images (last image of first row ADE the percentage of the head boundary that is not covered by the and the three images in the second row) belong to case study 3, poses 1, 2, 3 area of the projected skull is computed as a measure of the quality and 4, respectively. of the overlay. Figure 2 shows an example of the application of this evaluation procedure. The second real-world case considered corresponds to a Moroccan woman whose corpse was found in the South of Spain. There is a single available photograph corresponding to that one in the alleged passport. In this case of study, the forensic experts identified 16 cephalometric landmarks following an imprecise approach (second photograph of Figure 3). Finally, The third case study happened in Cádiz, Spain. The four different photographs were provided by the relatives, which acquired them at different moments and in different poses and conditions. Hence, this case study consists of four distinct SFO

problem instances. The forensic experts were able to locate 14, 16, Fig. 2. Example of area deviation error procedure. 15 and 8 landmarks following an imprecise approach for poses 1, 2, From left to right, original photographs (top) and projected skull (bottom), 3 and 4, respectively (third to sixth photograph of Figure 3). intermediate images with the head boundary (top) and binary skull (bottom), 5.3. Coevolution vs fuzzy landmarks and final XOR image (right most) with the corresponding ADE value below Table 1 shows the run time of both approaches in every SFO the image. instance. Our coevolutionary proposal is, between, ten and thirty

times faster than the fuzzy-evolutionary method. As said, these big

differences arise mainly because of the high computational effort 5.2. Case studies needed to calculate the fuzzy distance between each pair of As said, all the SFO instances tackled in this work are real-world crisp-fuzzy landmarks in every fitness evaluation. cases previously addressed by the Physical Anthropology lab at the Table 2 presents the ADE values for the obtained SFOs in the University of Granada (Spain). We briefly introduce them as previous six cases, distinguishing between fuzzy-evolutionary and follows. coevolutionary approaches. The minimum (m), maximum (M), The first case study happened in Málaga, Spain. The facial mean (ϖ), and standard deviation (σ) values of the thirty runs photograph of this missing lady was provided by the family and the performed are shown for each case. final identification done by CS has been confirmed. The forensic Mean and standard deviation ADE values show a better experts manually selected a set of fifteen 2D cephalometric performance of the fuzzy-evolutionary approach. However, landmarks on the face present in the photo, following an imprecise

24 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

according to the minimum ADE results, the coevolutionary obtained following both approaches. In fact, visual differences can approach outperforms in four of the six cases. The corresponding be mainly distinguished for those cases with bigger ADE SFO representation are depicted in Figure 4. As ADE values differences, that is, cases 1 and 3 poses 3 and 4. confirm, there are not big visual differences among the best SFO

Fig. 4. Best superimposition obtained following a fuzzy-evolutionary (first row) and a coevolutionary (second row) approaches. From left to right, case studies 1, 2, 3 (poses 1 to 4).

Table 1 Mean run time needed to execute the fuzzy-evolutionary and the coevolutionary approaches over each of the SFO instances. Run time Approach Case 1 Case 2 Case 3, 1 Case 3, 2 Case 3, 3 Case 3, 4

Fuzzy 55’ 190’ 63’ 82’ 97’ 210’

Coevolution 1.8’ 4.8’ 3.9’ 2.7’ 3.2’ 4.6’

Table 2 Mean run time needed to execute the fuzzy-evolutionary and the coevolutionary approaches over each of the SFO instances. ADE Case Approach m M ϖ σ

1 Fuzzy 15.20 45.98 21.56 10.60

Coevolution 12.20 22.16 17.44 2.74

2 Fuzzy 14.16 15.98 14.97 0.43

Coevolution 13.04 25.23 17.23 3.37

3, 1 Fuzzy 14.54 20.65 16.50 1.14

Coevolution 15.42 41.92 30.19 7.11

3, 2 Fuzzy 23.67 26.07 24.78 0.57

Coevolution 23.59 37.71 31.24 3.84

3, 3 Fuzzy 17.32 20.62 18.97 0.88

Coevolution 14.22 38.55 26.57 6.58

3, 4 Fuzzy 16.70 18.14 17.38 0.36

Coevolution 19.14 34.36 28.22 3.37

25 Óscar Ibáñez, Óscar Cordón, Sergio Damas. Two different approaches to handle landmark location uncertainty in skull-face overlay: Coevolution vs fuzzy landmarks. Forensic Sci Sem, 2018, 8(1): 19-26.

6. CONCLUDING REMARKS AND FUTURE WORKS 2008. 5. O. Ibáñez, L. Ballerini, O. Cordón, S. Damas, and J. Santamaría. An experimental study on the applicability of evolutionary algorithms to In this paper we have proposed a coevolutionary approach to deal craniofacial superimposition in forensic identification. Information with imprecise cephalometric landmarks in the SFO process. By Sciences, 179:3998–4028, 2009. using a CCEA we are able to look for both the best transformation 6. O. Ibáñez, O. Cordón, S. Damas, and J. Santamaría. A new approach to fuzzy location of cephalometric landmarks in craniofacial parameters and the best landmark locations at the same time. superimposition. In Proc. of the International Fuzzy Systems Our proposal has been tested on six complex real-world Association - European Society for Fuzzy Logic and Technologies (IFSAEUSFLAT) World Congress, pages 195–200, Lisbon, Portugal, identification cases and it has been compared with our previous 2009. fuzzy-evolutionary method. CCEA is a promising approach for 7. O. Ibáñez, O. Cordón, S. Damas, and J. Santamaría. Modeling the skull-face overlay uncertainty using fuzzy sets. Technical Report SFO since it achieves outstanding results in very short time AFE 2010-06, European Centre for Soft Computing, Mieres, Spain, (between ten and thirty times faster than our previous method). 2010. Submitted. 8. J. Paredis. Artificial Life, 2:355–375, 1995. Furthermore, the best coevolutionary results are better than the 9. B. Zitová and J. Flusser. Image registration methods: a survey. Image fuzzy-evolutionary ones in four of the six problem instances. and Vision Computing, 21:977–1000, 2003. However, the robustness of the method should be increased. 10. D. Hearn and M. Pauline Baker. Computer graphics (2nd ed.): C version. Prentice-Hall, Inc., Upper Saddle River, NJ, USA, 1997. With that aim, we are studying new ways of interaction between 11. T. Bäck. Evolutionary Algorithms in Theory and Practice: Evolution the two populations. Strategies, Evolutionary Programming, Genetic Algorithms. Oxford University Press, 1996. 12. N. Hansen and A. Ostermeier. Adapting arbitrary normal mutation ACKNOWLEDGMENTS distributions in evolution strategies: The covariance matrix adaptation. In Proceedings of the 1996 IEEE International This work is supported by the Spanish Ministerio de Educación y Conference on Evolutionary Computation, pages 312–317, Ciencia (ref. TIN2009-07727), including EDRF fundings. We Piscataway, New Jersey, 1996. 13. F. Herrera, M. Lozano, and J. L. Verdegay. Tackling real-coded would like to acknowledge all the team of the Physical genetic algorithms: operators and tools for the behavioural analysis. Anthropology lab at the University of Granada (headed by Dr. Artificial Intelligence Reviews, 12(4):265–319, 1998. 14. J. Richtsmeier, C. Paik, P. Elfert, T.M. Cole, and F. Dahlman. Botella and Dr. Alemán) for their support during the data Precision, repeatability and validation of the localization of cranial acquisition and validation processes. Part of the experiments landmarks using computed tomography scans. The Cleft related to this work was supported by the computing resources at Palate-Craniofacial Journal, 32(3):217–227, 1995. 15. J. Santamaría, O. Cordón, S. Damas, and O. Ibáñez. Tackling the the Supercomputing Center of Galicia (CESGA), Spain. coplanarity problem in 3D camera calibration by means of fuzzy landmarks: a performance study in forensic craniofacial superimposition. In IEEE International Conference on Computer Vision, pages 1686–1693, Kyoto, Japan, 2009. REFERENCE 16. D. Dubois and H. Prade. On distance between fuzzy points and their use for pausible reasoning. In International Conference on Systems, 1. W. M. Krogman and M. Y. Iscan. The human skeleton in forensic Man and Cybernetics, pages 300–303, 1983. medicine. Charles C. Thomas, Springfield, IL, 1986. 2nd edition. 17. I. Bloch. On fuzzy distances and their use in image processing under 2. M. Y. Iscan. Introduction to techniques for photographic comparison. imprecision. Pattern Recognition, 32:1873–1895, 1999. In M. Y. Iscan and R. Helmer, editors, Forensic Analysis of the Skull, 18. R. P. Wiegand. An Analysis of Cooperative Coevolutionary pages 57–90. Wiley, 1993. Algorithms. PhD thesis, George Mason University, Fairfax, Virginia, 3. C. N. Stephan. Craniofacial identification: techniques of facial 2003. approximation and craniofacial superimposition. In S. Blau and D. H. 19. M. A. Potter and K. A. De Jong. Evolutionary Computation, Ubelaker, editors, Handbook of Forensic Anthropology and 8(1):1–29, 2000. Archaeology, pages 304–321. Left Coast Press, California, USA, 20. Liles W. Wiegand, R. P. and K. De Jong. An empirical analysis of 2009. collaboration methods in cooperative coevolutionary algorithms. See 4. T. W. Fenton, A. N. Heard, and N. J. Sauer. Skull-photo Spector, Goodman, Wu, Langdon, Voigt, Gen, Sen, Dorigo, Pezeshk, superimposition and border deaths: identification through exclusion Garzon, and Burke, pages 1235-1242, 2001. and the failure to exclude. Journal of Forensic Sciences, 53(1):34–40,

POSTSCRIPT First reported on the EUSFLAT-LFA 2011, Aix-les-Bains, France, July, 2011.

26 Hongfa Chang, Daiqin Tao, Shengxiong Liu, et al. Deceleration impact experiments and deceleration injury mechanism analysis on the thoracic and abdominal organs. Forensic Sci Sem, 2018, 8(1): 27-31.

Peer Reviewed Forensic Science Seminar Received 26 April 2013 ISSN 2157-118X Received in revised form 9 March 2018 Volume 8 Number 1 Accepted 2 April 2018 3 April 2017 Available online 3 April 2018

eceleration Impact Experiments and DDeceleration Injury Mechanism Analysis on the Thoracic and Abdominal Organs

Hongfa Chang, M.S. a, Sn. Engr., Dr., Daiqin Tao, Ph.D. a, b, *, Assoc. Prof., Dr., Shengxiong Liu, Ph.D. b, c, Jingru Xie b, Dr., Guangyu Yang, Ph.D. b, Prof., Dr., Zhiyong Yin, Ph.D b, c, Academician CAE, Dr., Zhengguo Wang, Ph.D. b, c

a. Chongqing Engineering Research Center for Criminal Investigation Technology, Institute of Forensic Science, Criminal Police Corps of Chongqing Public Security Bureau, Chongqing 400707, China b. Chongqing Key Laboratory of Vehicle/Biological Crash Security, Research Institute for Traffic Medicine, Institute of Surgery Research, Daping Hospital, Third Military Medical University, Chongqing 400042, China. c. School of Pharmacy & Bioengineering, Chongqing University of Technology, Chongqing 400054, China *CORRESPONDING AUTHOR. E-mail: [email protected]. Tel.: +86-023-63964932

ABSTRACT Objective It takes animal organs as models to study the process and mechanisms of thoracic and abdominal organs deceleration injury to establish a more reasonable basis for the injury assessment. Methods New Zealand rabbits were dissected along the cross-section of the chest near the diaphragm as animal models. The chest and organs were exposure. The supine (Group A) and the prone (Group B) positions were designed to be fixed in the crash test platform. The movable platform free fell from fixed height and hit the fixed platform. The high-speed camera was used to record the whole deceleration impact process. After completion of the experiment, the image sequence analysis software was used to analyze the collision process. Results The injured areas were all at the impact side. The high-speed camera showed that before ribs squeezing the lung at the hedge side, organs had hit the chest wall at the impact side then compressed. When ribs deformed at the hedge side, they did not compress organs immediately but with a time interval before contact and compression. The deformation of Group A was significantly higher than Group B, but there were no significant differences for the severity of organ damages. Conclusions In deceleration injury, the thoracic and abdominal organ damage is mainly due to the inertia impact between organs and the chest wall, while it is of no direct correlation with the compressional deformation. Thus, the AIS injury assessment criteria have serious problems, and it needs to find a new scientific assessment method. KEY WORDS Injury, Impact, Liver, Deceleration impact, Animal experiment, High speed camera, Forensic science

Severity Score (ISS) is an established medical score to assess 1. INTRODUCTION trauma severity, and the current damage assessment is using the When the vehicle encountered obstacles in the process of moving, AIS standard related to ISS [1, 2]. It correlates with mortality, it could be involved in a sharp slowdown with large probability. morbidity and hospitalization time after trauma. It is used to define For the fighter, it also needs a sharp slowdown process in the the term major trauma. And the Abbreviated Injury Scale (AIS) is moment of landing for various venues. In these deceleration an anatomically based consensus-derived global severity scoring situations, abdomens of occupants inside the aircraft / vehicle may system that classified each injury in every body region according to have a powerful contact collision with the machine parts causing its relative severity on a six-point ordinal scale: 1 Minor, 2 organ damage. Therefore, it is necessary to study the biomechanics Moderate, 3 Serious, 4 Severe, 5 Critical, 6 Maximal (currently mechanism of organ injury generated by the deceleration impact. untreatable). As a reflection of existing research results in this field, the Injury Early studies have discovered that organs in the chest and

27 Hongfa Chang, Daiqin Tao, Shengxiong Liu, et al. Deceleration impact experiments and deceleration injury mechanism analysis on the thoracic and abdominal organs. Forensic Sci Sem, 2018, 8(1): 27-31.

abdomen are so fragile and of little elasticity that they are very [5]. easily to be injured during the impact [3]. And there have been New Zealand white rabbits were anesthetized and then fixed in several preliminary conclusions [4] for these organs injury the experimental platform. After causing them the suffocation death, mechanism: 1) During the deceleration impact, organs encounter their abdominal cavities were opened. From the 9th spine, the more substantial squeeze between the anterior abdominal wall and thoracic cage was cut off and the head was cut. The diaphragm was the spine; 2) Viscous injury: due to over-stretch effects and shear removed and the lungs were exposed. Placed the chest and organs effects caused by the increased fluid pressure in the liver on high on the platform of the deceleration device and fix it. Group A: chest load; 3) In sudden deceleration, the liver does the inertia movement plate, back up (Figure 1). Group B: Back plate with chest facing up that stretch lacerations occurr at the site of attachment of blood (Figure 2). vessels and ligaments; 4) Ribs occur stabbed at the end of the fracture. For organs in case of sudden deceleration, their own injury, which is having nothing to do with the external connection but caused by inertial characteristics, is rarely reported in the previous literature. This paper will preliminary study the mechanical effect caused by its inertia characteristics in organs reducer impact process. Relations among the AIS, the rib fracture and the chest compression:

Chest Compression → Rib Fracture → AIS AIS = -3.78 + 0.1956 * Chest Compression (%) Fig. 1. Group A: Thoracic cross section (before impact) Since it is through the classification assessment to determine the injury level, AIS is mainly based on and developed through the dummy experiments But the dummy is a frame structure. It is a cavity and supported by a bone structure with no soft issues such as organs inside. So, for this method, it can only assess the inner organ damages by the compression degrees of the thoracic and abdominal wall. For human body, in traffic accidents, each part is moving at the same speed. But when the vehicle stopped, the body in the vehicle is still moving forward due to the inertia. So different parts of the body have different speeds. When the chest hits the vehicle, organs are still moving forward and then impact with the chest wall. Fig. 2. Group B: Thoracic cross section (before impact) Because of the resistance from the rib profile, the extrusion slows

done. Thus, the organ impact with the back should be later than

with the chest wall. Took out the lungs and heart, took pictures, and observed. For AIS, it does not consider the injury differences between Finally, the image sequence analysis software was used to deceleration and acceleration. So, it needs to do experiments to analyze the injury condition. study the real mechanism. It should make clear whether the organ deceleration injury is caused from the chest wall impact by the 3. RESULTS inertia force or from the back compression by the deformation. Both groups of lungs collapsed, but no significant changes were In this research, rabbits are taken as the animal model for seen in the kidneys. observing the injury process. The organ impact properties have also There was bleeding damage on the back panel of Group A been detected to establish a basis for more reasonable assessment (Figure 3). There was bleeding damage on the chest plate of Group on organ injury. B (Figure 4). Both groups were contact injured. 2. MATERIALS & METHODS For Group A, as shown in Figure 5, the frames under high-speed camera displayed that, during the deceleration process (the red line Materials：Two groups of New Zealand white rabbits with 9 per on Figure 5), due to the inertia of the organ, it will first collide with group and each weight 2.5 0.1kg (Animal Centre of Daping the spine, and then it will impact gradually (the point a on Figure 5) Hospital); a high-speed camera（PHANTOM V4.3, USA）; and the with the impact side of the rib wall (the yellow line on Figure 5), Biological Impact Device with Free-fall Deceleration Impact Rail and a series of blocked compression motions began. At this time,

28 Hongfa Chang, Daiqin Tao, Shengxiong Liu, et al. Deceleration impact experiments and deceleration injury mechanism analysis on the thoracic and abdominal organs. Forensic Sci Sem, 2018, 8(1): 27-31.

the front part of the chest (the blue line on Figure 5) was deformed For Group B, as shown in Figure 6, the frames under high-speed without constricting the organs. Until the deformation speed of the camera displayed that, during the deceleration process (the red line anterior chest synchronously exceeds the obstructed compression on Figure 6), due to the inertia of the organ, it will first collide with velocity of the organ, the organ is subjected to the deformation the front part of the chest, and then it will also impact gradually force of the chest cavity (the point b on Figure 5). The organ was (the point a on Figure 6) with the impact side of the rib wall (the affected by the inertial force (the point a on Figure 5) before the rib yellow line on Figure 6), and a series of blocked compression compression force (the point b on Figure 5). motions began.

Fig. 3. Group A: After impact Fig. 4. Group B: After impact

Fig. 5. Group A: High-speed camera shots during the crash

Fig. 6. Group B: High-speed camera shots during the crash

At this time, the back of the chest (the blue line on Figure 6) was was affected by the inertial force (the point a on Figure 5) before deformed without constricting the organs. Until the deformation the rib compression force (the point b on Figure 6). speed of the anterior chest synchronously exceeds the obstructed According to the high-speed video time series frame diagram, compression velocity of the organ, the organ is subjected to the the deformation in Group A was significantly higher than that in deformation force of the back (the point b on Figure 6). The organ Group B, but there was no significant difference in the severity.

29 Hongfa Chang, Daiqin Tao, Shengxiong Liu, et al. Deceleration impact experiments and deceleration injury mechanism analysis on the thoracic and abdominal organs. Forensic Sci Sem, 2018, 8(1): 27-31.

4. DISCUSSION complex research direction. The results of the current study are mixed with far-off non-serialized and standardized data such as the The experimental results show that: During the deceleration and impact experiment of human fresh entities, and the injury impact of the liver, a significant laceration occurred on the opposite experiment of anesthetized animals caused by seat belts. The shape, side of the impact, and no obvious tearing injury was found on the area, impact velocity, and impact mode of the impactor can impact side. Although the severity of the injuries was different due undergo complex changes, leaving many gaps to be explored. [4] to the individual differences in animal tissues, the results still Due to ethical constraints, the liver impact injury experiment suggest that the opposite side was subjected to a pronounced tensile cannot obtain sufficient data from human morbidity experiments. stress; and the impact stress on the impact side was relatively weak. Therefore, it is necessary to carry out extensive research on new The results of studies on the mechanism of hepatic impact injury experimental methods, experimental models, and alternatives. have been summarized in the past as crush injury, stretch injury and With the appearance of new experimental instruments and test shear injury caused by increased intrahepatic fluid pressure, methods, including the multi-functional small biological impact traction tears of peripheral blood vessels and ligaments, and stab machine developed by Prof. Zhiyong Yin and others [7], as well as wounds. [4] The mechanism of stretch injury caused by increased the emergence of various biological substitutes or simulants, it intrahepatic fluid pressure is different from that of this experiment. provides good conditions for biomechanical research. Biological The former is mainly due to the explosive stretch injury caused by substitutes have similar organizational structure and mechanical a sudden increase in the pressure of the intrahepatic fluid caused by properties to human organs, and also have the ability to reflect the impact of the liver in vivo in the presence of liquid filling. In different degrees of damage. They can be used under different this experiment, the liver was obtained after the animals were impact conditions (impact direction, form, impact velocity, loading sacrificed. There is no liquid filling condition in the liver. Therefore, mode, and some living organisms cannot be achieved. (Conditional) the tear damage generated in the experiment should not be due to a biomechanical studies. These biological substitutes or mimics can sudden increase in intrahepatic fluid pressure. In this experiment, provide meaningful suggestions for studying the protective the injury mechanism of the liver should be that during the impact, principles of liver injury. the liver is in the force field due to its own inertia and impact The thoracic frame formed by the ribs and spine is anteriorly soft reaction force, and there is only a constraint between the tissue and and hard. Their compression is not the same. If it is squeezed, the the tissue in the heel-side tissue, so the opposite side is stretched. injury before and after it should be different. However, the Stress can easily cause lacerations. The impact side did not see a experimental results show that under the opposite placement mode, significant laceration during the impact. This is due to the fact that the deceleration impact of the visceral organs corresponding to the the impact-side tissue is not only bound by tissue-tissue constraints, spine parts is the same (all with the same degree of injury on the but also because the impact-side tissue is in close contact with the impact surface). This shows that organ injury is not thoracic impact contact surface, and the interface between different media compression, but mainly inertial collisions. can produce a certain amount of friction thus hinders mutual

displacement between different media. This boundary condition

considerably weakens the stretch effect on the impact side of the liver, so that the tear damage due to stretching can be reduced to a corresponding extent. In actual traffic accidents, there are two mechanical mechanisms for the tearing of the human liver: The first is due to the sudden increase in intrahepatic fluid pressure during the impact of the impact of the liver bursting tear injury occurs, and its mechanical action is from the inside out; the second is the impact of the heel side of the liver stretched the role of stress caused by its tear damage, and its mechanical action is from outside to inside. For the dose-effect relationship between impact energy and [6] injury level in liver impact injury, relevant literature has been Fig. 7. Effect of deceleration process on organs (viscera) reported in detail. This experiment focuses on the study of the mechanism of injury in the process of hepatic impact injury. It is expected to help to clarify the mechanical mechanism of liver In the case of squeezing deformation, there should be traces of injury that is common in traffic accident injuries to a certain extent, damage on both the upper and lower organs, but in the experiment, and to provide some new experimental methods and certain there were only traces and no traces. Accelerated injuries are Experimental basis. caused by squeezing, such as damage caused by a blow, but the The biomechanical mechanism of liver injury is a very broad and

30 Hongfa Chang, Daiqin Tao, Shengxiong Liu, et al. Deceleration impact experiments and deceleration injury mechanism analysis on the thoracic and abdominal organs. Forensic Sci Sem, 2018, 8(1): 27-31.

reverse is the opposite (Figure 7), such as damage caused by a car establishment of a more rational method of chest and abdomen accident. Due to the difference in force of the type of impact, the organ damage assessment of injury mechanisms. shape of these two types of injuries is very different. ACKNOWLEDGMENTS Inertial injury is only damage to the contact surface. The crush injury is symmetrical, and it should be that both the contact surface This paper is supported by the Natural Science Foundation of and the corresponding surface are damaged. China (NSFC31200709), the Major Research Projects of the 13th In previous studies on road traffic injuries, many researchers Five-Year Plan of the Ministry of Justice of the People's Republic believed that occupant injury to the thoracic and abdominal organs of China (2016YFC0800702 and 2016YFC0800702-3). The was caused by external forces acting on the ribs, which in turn authors would like to acknowledge the constructive comments caused internal organs to squash and cause injuries. The indicators given by the anonymous reviewers. including chest and abdomen in the assessment of existing vehicle safety performance are still based on the amount of chest REFERENCE compression. The results of this experiment show that the damage 1. Baker, S. P.; B. O'Neill, W. Haddon Jr., W. B. Long (1974). "The to the thoracic organs during deceleration injury is mainly caused Injury Severity Score: a method for describing patients with multiple injuries and evaluating emergency care". The Journal of Trauma by the damage of the organs against the chest wall. Therefore, there (Lippincott Williams & Wilkins) 14 (3): 187-196. doi: are serious problems in the AIS damage assessment criteria and 10.1097/00005373-1974030000-00001.PMID 4814394. 2. Copes, W. S.; H. R. Champion, W. J. Sacco, M. M. Lawnick, S. L. new scientific evaluation methods need to be found. Keast, L. W. Bain (1988). "The Injury Severity Score revisited". The Journal of Trauma (Lippincott Williams & Wilkins) 28 (1): 69-77. 6. CONCLUSIONS Doi: 10.1097/00005373-198801000-00010. PMID 3123707. 3. Cooper GJ, Taylor DE.Biophysics of impact injury to the chest and Due to the development of computer technology and the rise of abdomen[J].J R Army Med Corps,1989,135(2):58-67. 4. Zhengguo Zheng. Modern Communications Medicine [M]. 1st Ed. the artificial intelligence wave, some research teams in the Chongqing: Chongqing Publishing Group & Chongqing Publishing academic community are currently working on relevant research House, 2011: 256-262, 585-596. through computer modeling, virtual simulation, and intelligent 5. Zhao Hui, Yin Zhiyong, Ning Xin, et al. Portable multifunctional biological impact machine and method for obtaining impact energy analysis. However, the premise of all virtual technologies is still based on the impact machine: China, CN201310408039.2 [P]. inseparable from the verification of real data. This experiment has 2013-12-25. (In Chinese) 6. LU Ming, TANG Li-jun, TIAN Fu-zhou, et al. Establishment and shown that, in the deceleration injury, the thoracic and abdominal evaluation of a liver trauma model in rats [J]. J Trauma Surg, 2010, organ damage is mainly caused by impact due to the organ inertia, 12(1): 60-63. (In Chinese) 7. Yin Zhiyong, Wang Zhengguo, Liu Haipeng, et al. The development and it is of no direct correlation with the compressional and application of a minitype multifunctional bi-impactor [J]. J deformation. The conclusion is of great significance for the Biomed Eng, 2000, 17(3): 309-312. (In Chinese)

POSTSCRIPT N/A

31 Mely Purnadianti, Andika Aliviameita, Diah Ayu Nur Rochmawati, et al. Effects of heat exposure duration on saliva traces on cigarette butts as forensic identification tools. Forensic Sci Sem, 2018, 8(1): 32-34.

Peer Reviewed Forensic Science Seminar Received 13 October 2017 ISSN 2157-118X Received in revised form 5 December 2017 Volume 8 Number 1 Accepted 5 December 2017 3 April 2017 Available online 3 April 2018

ffects of Heat Exposure Duration on Saliva Traces Eon Cigarette butts as Forensic Identification Tools Mely Purnadianti a, *, Andika Aliviameita a, Diah Ayu Nur Rochmawati a, Dian Amanovitasari a

a. Airlangga University, Surabaya, Indonesia *CORRESPONDING AUTHOR: Master of Forensic Science. Addr.: Ilmu Forensik, Sekolah Pascasarjana, Universitas Airlangga Surabaya, Jl. Airlangga 4-6, Surabaya, Indonesia. E-mail: [email protected]. Tel.: +6282136634060

ABSTRACT Identification processes are not only conducted on the dead bodies of criminal victims, but also on the evidence found at the location. The length of time between the crimes happening and the investigation of the evidence is a problem in relation to the blood group identification process through saliva traces on cigarette butts. Wet saliva usually becomes trace in an hour and 40 minutes at room temperature, and up to less than 3 hours if there any influencing factors. The purpose of this research is to analyze the effects of heat exposure duration on the protein content of the saliva traces on cigarette butts in order to help the forensic blood group identification process. A time series design was used in this research in which 18 cigarette butts were collected from 6 individuals of blood groups A, B, and AB. The samples were then incubated for 1.3 and 6 hours respectively. The protein content was determined using a Trizol reagent and Spectrophotometer UV. Statistical analysis using a T-test was conducted to analyze the results. There was a decrease in the protein content of the saliva trace samples at the time points of 1, 3 and 6 hours. The protein content of the saliva could still be detected at the time points of 1, 3 and 6 hours. These findings suggest that the protein analysis of saliva trace samples on cigarette butts can be used in the forensic blood group identification process. KEY WORDS Cigarette butts, Saliva traces, Protein, Forensic identification, Forensic science

identification process through the saliva traces on cigarette butts. 1. INTRODUCTION Wet saliva usually become trace in an hour and 40 minutes at room Forensic identification is the methods undertaken by temperature, and up to less than 3 hours if there any influencing investigators to investigate the identity of the victims and culprits. factors (Jellinghaus et al., 2015). Early blood group identification The identification processes are not only conducted on the dead through saliva samples on cigarette butt evidence needs to be body of criminal victims, but also on the evidence found at the conducted in order to help investigate the identity of the victims location (Gani, 2002). In the murder case that happened on 14th and culprits. The aim of this study is to determine the time span for May 2013 in Bangkingan Madura, East Java, several items of blood group analysis through saliva trace samples down to the information like the smoking habits, brand, lip print, finger print remaining protein levels. and even saliva traces can be detected on the cigarette butt as 2. METHOD evidence found (Hardjanto Pudji, 2015). The saliva traces can be used as a specimen in blood group identification which is a This study was a laboratory experiment using a time series design secondary identification component (Rogers et al, 2005). The which used 1, 3 and 6 hours at room temperature as the respective glicoprotein content in saliva is a component used to determine intervals. The samples used in this study consisted of 18 cigarette blood group in the ABO and Lewis system in individuals who have butts. In this study, the saliva on the cigarette butts was set up as a secretor blood group (Albertolle et al., 2017). A research study the independent variable, while the dependent variable was the conducted at a State Police Laboratory in Jakarta found that 75% of identification of the forensic blood type. The control variable in police officers have a secretor blood group. On the other hand, 85% this study was the time duration used by the researchers that ranged of the world’s population have a non- secretor blood group. between 1 hour, 3 hours and 6 hours to determine the decreasing The length of time between crimes happening and the protein count in saliva. The confounding variable was the investigation of the evidence is a problem in the blood group temperature because it can lead to the saliva drying on the cigarette

32 Mely Purnadianti, Andika Aliviameita, Diah Ayu Nur Rochmawati, et al. Effects of heat exposure duration on saliva traces on cigarette butts as forensic identification tools. Forensic Sci Sem, 2018, 8(1): 32-34.

butts. This study was conducted in the TDC (Tropical Disease The protein level of the cigarette butts in Table 6 shows that after Centre) Laboratory at the Tropical Disease Centre Institute of one hour post-smoke, the swab ranged from 0.686 mg/ml to 2,958 Airlangga University, Surabaya and the forensic laboratory of mg/ml, after 3 hours post-smoke the swab ranged from 1.355 Saiful Anwar Public Hospital, in Malang. A non-parametric statistic mg/ml to 1.818 mg/ml, and after 1 hour post- smoke the swab test (normality test, homogeneity, t test) with linear regression ranged from 0.564 mg/ml to 1.839 mg/ml. This was influenced by analysis was used to determine the influence of the exposure time the saliva spotted surface area of the subjects on the cigarette butts to the temperature with the amount of the protein level in the saliva and the salivary degree due to hydration. The greater surface area on the cigarette butt. of the salivary spots and its corresponding hydration level, the greater the salivary protein that is obtained. Table 1 shows the 3. RESULTS AND DISCUSSION average protein level at each swab time interval on the cigarette The samples were obtained from six cigarette butts smoked by six butt. The data shows a decreasing protein level over the exposure individual research subjects who know their blood type. The duration at room temperature. Figure 1 shows the decreasing curve cigarette butt with sample number 1 (A1 - F1) was swabbed one of the protein level. There was a significant decrease in the hour after being smoked; the cigarette butt with sample number 2 protein level of the 3rd swab compared to the 6th swab. (A2 - F2) was swabbed three hours after being smoked; the The protein level data of each exposure time group had a normal cigarette butt with sample number 3 (A3 - F3) was swabbed six distribution with a significance value of >0.05. In the homogeneity hours after being smoked. The estimation of the protein level in test using a T test, the significance value was 0.086 (>0.05), which saliva was done using a Spectrophotometer using a 280 nm means that there were at least two groups where there was no wavelength, which used an absorbance ratio of 280/260 to significant difference in the protein level. In the Post Hoc test, the determine the correction factor that exists in the table. The protein homogeneity test was performed between the groups of exposure level is determined by the following formula: protein level (mg / ml) time; a significant difference was found between the group with 1 = A280 x correction factor x dilution. hour of exposure time and the 6-hour group with a significance value of 0,038 (<0,05). The other tests do not show a significant difference with a significance value of > 0.05. The correlation test Table 1 Average protein level. between exposure time at room temperature and the protein levels using a Pearson correlation test, shows a significance value of No. Exposure duration at room Mean protein content 0.013 (<0.05). This means that there is significant correlation temperature (hours) (mg/ml) between exposure time at room temperature and salivary protein 1 1 1,43 levels. The correlation value is -0.499, indicating that the longer the 2 3 1,49 exposure time, the smaller the level of salivary protein. The 3 6 1,86 correlation strength is moderate.

4. CONCLUSION

The exposure time at room temperature has been shown to influence the levels of protein in the saliva on cigarette butts. There

is a decreasing level of protein in the salivary swab at 1 hour, 3

hours and 6 hours after being smoked with an average level

1.717969 mg/ml, 1.673408 mg/ml, 1.113023 mg/ml, and 0.988335

mg/ml respectively. Statistically, there is a significant correlation

between the duration of the exposure time at room temperature with salivary protein level. Moreover, the correlation value shows that the longer the exposure time, the smaller the level of salivary protein, the correlation strength is moderate. The mean value of the protein content of the salivary swab 1 hour after bite with the 6 3 1 0 salivary swab 6 hours after bite was also obtained.

Waktu per 1 jam REFERENCE

1. Afifah, Ratna, Nurul. Pretrial and Scope, Akademika Pressindo, Jakarta, 1986. Fig. 1. Curve of Average Protein Level - Exposure Duration at Room 2. Albertolle, Matthew E; Hassis, Maria E; Ng, Connie Jen; Et Al. Mass Temperature Spectrometry-Based Analyzes Showing the Effects of Secretor and Blood Group Status on Salivary N-Glycosylation. Clinical Proteomics Journal.2015. [Retrieved January 2017.]

33 Mely Purnadianti, Andika Aliviameita, Diah Ayu Nur Rochmawati, et al. Effects of heat exposure duration on saliva traces on cigarette butts as forensic identification tools. Forensic Sci Sem, 2018, 8(1): 32-34.

3. Alfanie Iwan, Nirmalasari Nila, Arizal Hendi Muhammad. Forensic Toxicology. International Journal of drug testing. P: 1-35, 1999 and Medicolegal Medical Sciences. Jakarta: Rajawali Pers, 2017. [Accessed February 2017] 4. Asthia T. Comparative Study of Blood Type Examination Result on 13. Idries. The Application of Forensic Medicine in the Investigation Saliva Sample Immediately and Saliva is stored for 1 hour at Process. Jakarta: Sagung Seto, 2008. Temperature 15º C. Thesis. Universityof Indonesia. Jakarta, 2007. 14. Jellinghaus K, Blasser L, Scheller C Bohnert M. Drying time for 5. De Almeida, Et Al. Saliva Composition and Functions: A Human Saliva. Arch Criminal Mart-Apr; 235 (3-4): 99-109. 2015. Comprehensive Review. The Journal of Contemporary Dental https: //www.ncbi.nlm.nih.gof/pubmed/26419084. [Accessed Practice.2008. [Accessed January 2017.] February 2017.] 6. Fawles, J et al. The Chemical Constituent In Cigarettes And Cigarette 15. Joachim Klimek. Saliva and Oral Health. Lecture Handout for Smoke. New Zealand: New Zealand Ministry of Health. 2000. Undergraduate Student of Dentistry. 2004 Germany: 4-37 [Retrieved February 2017.] 16. Leffondre et al. Modeling Smoking History. American Journal of 7. Gani. Forensic Medicine. Faculty of Medicine Andalas University. Epidemilogy, 2002. Padang, 2002 Http://aj.oxford.journal.org/contant/156/9/813.full.pdf. [Accessed 8. Gerard J. Tortora, Bryan Derrickson. Principles Of Anatomy And February 2017] Physiology 912th Edition. USA; John Wiley and Sons, Inc. 2009. 17. Martiman Prodjohamidjojo. Comment on Criminal Procedure Code: [Retrieved January 2017.] Criminal Procedure Code, Jakarta: Pradnya Paramitha, 2009. 9. Guyton and Hall. Book of Medical Physiology Edition 11. Jakarta: 18. Michael Bowers and Gary Bell. Journal Of Indian Academy Of Oral EGC, 2008 Medicine And Radiology, July-September 2011, 1995. [Accessed 10. Hardjanto Pudji. TK Talk. Surabaya: PT. Refka Petra Media, 2015. February 2017.] 11. Haque, amdadul M; Adhikari, Benu. Drying and Denaturations of 19. Tery Martin. Harmful Chemicals in Cigarettes. Journal of smoking proteins in spray drying process. Handbook of industrial drying. effect, 2012. http://quitsmoking.about.com [Discounted to february Taylor and francis Group, LLC, 2015. [Accessed January 2017.] 2017.] 12. Hold. K.M, Boer. D, Juidema. J. Saliva as an analytical tool in

POSTSCRIPT First reported on the 1st International Conference Postgraduate School Universitas Airlangga: Implementation of Climate Change Agreement to Meet Sustainable Development Goals (ICPSUAS 2017)

34 ERRATA: CORRIGENDUM & APOLOGY. FORENSIC SCI SEM, 2018, 8(1): C3. FORENSIC SCI SEM

ERRATA: CORRIGENDUM & APOLOGY

EDITORIAL *, a a Editorial Department, Forensic Science Seminar, Brooklyn, New York 11220, United States of America. * EDITORIAL. Email: [email protected]

There once had been a mistake in the publication of a wrong document, which has been corrected. Because of some mistakes in publishing, the most content of Professor James Byrne's paper (Ref.4.) had been wrongly typesetting and published online at Forensic Sci Sem, Vol. 7 No. 2., while the original correct article failed to be published in time. Since Professor James Byrne's paper (Ref.4.) is the only English reference for that article. It had been wrongly typed as the text and had not gotten valid proofreading. The wrong published document has been revoked and it has already published the correct manuscript online. About this mistake, we have already replied to the Taylor press. Authors of Forensic Sci Sem, Vol. 7 No. 2, who have downloaded the wrong version on the FSS official website, please go to the following address and download the corrected version. Please forgive us for the inconvenience for you. Download link for Forensic Sci Sem, Vol. 7 No. 2: http://fss.xxyy.info/journal/2017/2157118X.7.2.pdf

We apologize to Professor James Byrnethe again.

Ref.4. Reid, Joan A., and Eric Beauregard. 2015. "A Mixed Methods Exploratory Examination of Victim Injury and Death: Effect of Weapon Type and Victim Resistance during Sexual Assaults by Strangers." Victims & Offenders:1-24. doi: 10.1080/15564886.2015.1033580.

© 2018 by ZolCat Academic House. All rights reserved. No part of this journal may Brooklyn, NY 11220 be reproduced, in any form or by any means, E-mail: [email protected] without permission in writing from the publisher. http://fss.xxyy.info/ Printed in the United States of America ISSN 2157-118X 10 9 8 7 6 5 4 3 2 1 0

Forensic Science Seminar ( ISSN 2157-118X), VolumeDecember 8 number 2014 ISSN 1, April 2157-118X 3, 2018 C353 Forensic Science Seminar Volume 8 Number 1 April 2018 Single issue price $000USD (FREE)